article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
it is generally accepted that percolation is an essential aspect of gelation or vulcanization it is doubtful that even in a highly entangled melt of long polymers a nonzero value of the static shear modulus could exist in the absence of an infinite connected network .
however , percolation has usually been studied in rather special limits .
site and bond percolation of a single species on regular lattices are very well characterized and off - lattice percolation seems to present no new features @xcite , at least insofar as critical behavior is concerned .
more closely related to real gels are the so - called correlated percolation models where the distribution of crosslinks is drawn from a boltzmann distribution appropriate for a nearest neighbor lattice gas @xcite .
except at special points in the phase diagram these models are also in the universality class of the simple percolation problem . in our previous work on transport properties near the gel point @xcite ,
we have also used a simple one - species percolation process to produce the incipient gel .
we found that the shear viscosity diverges as the percolation concentration @xmath2 is approached according to @xmath5 with @xmath6 .
this value of the exponent @xmath4 is in excellent agreement with a prediction of de gennes based on a superconductor - normal conductor analogy @xcite and with recent analytical work on a rouse model @xcite .
it is also reasonably close to some experimental results for @xmath4 @xcite but quite different from that produced by another set of experiments @xmath7 @xcite .
thus it seems reasonable to ask if different versions of the crosslinking process might produce significantly different cluster size distributions from percolation and , consequently , different rheological properties .
gelation often occurs in the presence of a solvent and over some period of time rather than instantaneously , as in the usual percolation models . to simulate this feature ,
we have considered a two - species model consisting of a fraction @xmath8 of @xmath0-functional particles that are eligible to bond irreversibly to others of the same kind .
the remaining particles are inert and function as a background liquid through which the gel particles and clusters diffuse .
crosslinking occurs in stages : the equations of motion of all the particles are integrated forward for a fixed number of time steps between crosslinking attempts and this process is continued until the desired number of crosslinks is attained . at a critical concentration of crosslinks , @xmath2 , ( in the thermodynamic limit ) the largest cluster percolates and an amorphous solid forms . for this process one can calculate the usual static or geometrical quantities used to characterize percolating systems , _
e.g. _ , the fraction of particles on the ` infinite cluster ' , @xmath9 , the mean mass of finite clusters , @xmath10 , the fraction of samples percolating @xmath11 , the cluster size distribution @xmath12 where @xmath13 is the mass of a cluster and the radius of gyration @xmath14 where @xmath15 is the fractal dimension of the clusters . for simple percolation processes , @xmath16 , @xmath17 and
these two exponents determine the others through scaling relations @xcite . here
we find , at least for small @xmath8 , that the cluster size distribution , even at @xmath2 , is not well described by a simple power law .
however , the other static quantities listed above do display power law behavior near @xmath2 and a standard finite - size scaling analysis provides a very good collapse of our data .
moreover , the hyperscaling relation @xmath18 , where @xmath19 is the dimensionality and @xmath20 the correlation length exponent , is satisfied .
this suggests that this percolation transition is fundamentally describable in terms of a fixed point with two ( at least ) relevant scaling fields .
as the percolation point is approached from below , the shear viscosity diverges according to @xmath3 .
in contrast to our previous work on a model without solvent , we find values of @xmath4 in the range @xmath21 as compared with @xmath6 .
these results suggest that the critical behavior of transport coefficients of systems close to the gel point is nonuniversal .
the structure of this article is as follows . in section [ sec : model ]
we describe the present model and simulation procedures in more detail .
the geometric properties of the system are discussed in section [ sec : percolation ] and the data on the shear viscosity are presented in section [ sec : visco ] .
we conclude with a brief summary and discussion in section [ sec : discu ] .
we consider a system of @xmath22 particles in three dimensions , all of which interact with each other through the soft - sphere potential @xmath23 for @xmath24 @xcite with @xmath25 and @xmath26 .
we simulated systems at a volume fraction @xmath27 which is well below the liquid - solid coexistence density . in the absence of any other interactions ,
this system would be a simple three - dimensional liquid .
we initially place the particles on a simple cubic lattice that fills the computational box .
we then randomly select @xmath28 particles to be the gel forming component .
after equilibration of the system with brownian dynamics , with periodic boundary conditions , for 10000 time steps we begin the crosslinking process . at this point , the calculation proceeds via _ conservative _ molecular dynamics ( md ) so as to allow hydrodynamic modes to develop .
here we use a time step @xmath29 . in the smallest system ,
crosslinking is carried out one bond at a time .
a single gel particle is randomly selected and all other gel particles within a distance of @xmath30 are identified .
one of the particles in this list is randomly selected and bonded irreversibly to the central particle through the tethering potential @xmath31 with @xmath32 and @xmath33 .
each gel particle is allowed to bond to no more than six others and bonding between any pair of particles occurs at most once .
the configuration of the entire system is then updated for 100 time steps and the entire bonding process is repeated until @xmath34 crosslinks have been added .
the parameter @xmath1 is analogous to the occupation probability in a bond percolation process on the simple cubic lattice . in larger systems ,
the number of crosslinks added in the bonding steps is scaled by the system size in order to keep the crosslinking rate per gel particle constant .
the parameters in the potentials and the total volume fraction @xmath35 are the same as in our previous work @xcite .
the differences are that in this earlier work all particles were considered to be gel particles and that the crosslinking was done instantaneously , at @xmath36 , when the particles were on the vertices of a cubic lattice and thus all structural properties were those of percolation in three dimensions .
the present model is similar in some ways to a model discussed by gimel _
_ @xcite and hasmy and jullien @xcite who studied percolation in the context of diffusion - limited cluster - cluster aggregation using monte carlo methods .
their model differs from ours in that it is a lattice model , in the details of the crosslinking process , in the lack of solvent and in the nature of the cluster dynamics . in monte carlo simulations , one is forced to arbitrarily choose the mass - dependent diffusion constant @xmath37 whereas in our molecular dynamics calculations it is determined by the existing structure and the interparticle forces .
in the regime that is of interest here , _
i.e. _ , high enough gel density that percolation is possible , these authors find the critical behavior of ordinary percolation . in a separate set of runs ,
we calculate the stress - stress autocorrelation function and , through the appropriate green - kubo formula , the shear viscosity .
equilibration and crosslinking are carried out as described above and the calculation of the viscosity is again done with conservative md .
the adjustable parameters in our calculations are the gel fraction @xmath8 , the crosslink density @xmath1 and the system size . here
we report results for @xmath38 , @xmath39 and @xmath40 .
calculations for other values of @xmath8 are in progress and will be reported in a future publication @xcite .
we parametrize the size of our system in terms of the dimensionless length @xmath41 where @xmath22 is the total number of gel and solvent particles . because the crosslinking process is itself quite time consuming , we are able only to simulate systems up to size @xmath42 ( 32768 particles ) and this makes our estimates of critical exponents rather imprecise . a second factor contributing to the uncertainty in critical exponents
is that we need to determine the critical crosslink density @xmath2 for each value of @xmath8 whereas for lattice percolation this number is known to high accuracy .
we next discuss the static ( geometric ) properties of our model .
the critical concentration @xmath2 at which percolation occurs in the thermodynamic limit @xmath43 is accurately estimated from the intersection of curves @xmath44 as function of @xmath1 for different values of @xmath45 .
here @xmath44 is the fraction of samples percolating in a system of size @xmath45 at crosslink concentration @xmath1 .
for the two cases of interest here , @xmath46 and @xmath38 , we find @xmath47 and @xmath48 . once @xmath2 has been determined , the correlation length exponent @xmath20 can be estimated from the collapse of the data for the function @xmath0 when plotted as function of @xmath49 .
we show this collapse of the data for @xmath38 and @xmath39 in figure ( [ fig1 ] ) . for @xmath46 , the best collapse of the data for @xmath50
is obtained for @xmath51 which should be compared to the three - dimensional percolation result @xmath52 . for @xmath38 ,
finite - size effects are more pronounced and the data for @xmath53 have been excluded . for this case ,
the best collapse of the data is obtained for @xmath54 .
this method of estimating a critical exponent is not very accurate but the three - dimensional percolation value @xmath52 provides a significantly worse collapse of the data .
we next discuss the mean size of finite clusters because this data provides an unbiased estimate of the ratio @xmath55 . in the thermodynamic limit ,
@xmath10 with @xmath56 for @xmath19 percolation
. for finite @xmath45 , @xmath57 is peaked near @xmath2 with a peak height that grows as @xmath58 .
therefore , rescaling the peak heights to the same value for different @xmath45 provides an estimate of @xmath55 that is not affected by errors in either @xmath2 or @xmath20 . of course , the overall collapse of the data to a universal curve depends on accurate determination of these two quantities but the peak height does not . in figures ( [ fig2 ] ) and ( [ fig3 ] )
we show the function @xmath59 plotted as function of @xmath60 for the previously determined values of @xmath2 and @xmath20 .
the collapse to a universal curve is quite respectable for both @xmath46 and @xmath61 for @xmath62 and @xmath63 , respectively .
as above , the data for @xmath53 have been excluded for @xmath38 .
we note that in the case of three - dimensional percolation the ratio @xmath64 .
use of this value of @xmath55 in figure ( [ fig2 ] ) would result in a 40% difference between the peak heights for @xmath42 and @xmath53 . in the scaling theory of percolation @xcite , the ratio @xmath65 , where @xmath19 is the dimensionality and @xmath66 is the exponent characterising the cluster size distribution at @xmath67 .
if we enforce this scaling relation , we obtain @xmath68 for both @xmath38 and @xmath46 .
using @xmath69 , we find @xmath70 and @xmath71 .
using @xmath72 for the fractal dimension results in the prediction @xmath73 and @xmath74 for the fractal dimensions of the clusters .
as well , the hyperscaling relation @xmath75 yields @xmath76 and @xmath77 for @xmath46 and @xmath61 respectively . the accuracy of these scaling predictions is tested in figures ( [ fig4 ] ) to ( [ fig7 ] ) . in figure ( [ fig4 ] )
we show the number of clusters @xmath78 of mass @xmath13 at @xmath79 for @xmath38 and 0.3 for @xmath42 and @xmath80 . for the case @xmath81
, we have only counted the uncrosslinked gel particles . in neither case
is the data well described by a simple power law , in contrast to percolation on a lattice or in the absence of solvent where the exponent @xmath16 is already obtained for @xmath82 .
a fit to a power law over the range @xmath83 yields @xmath84 for @xmath38 and @xmath85 for @xmath46 .
the straight lines in figure ( [ fig4 ] ) are the best fits to the form @xmath86 over the range @xmath83 and while the fit is not perfect , the data are not inconsistent with this behavior in the limit of large @xmath13 . in figure ( [ fig5 ] )
we show the square of the radius of gyration @xmath87 as function of @xmath13 for a system of size @xmath42 together with curves @xmath88 with @xmath74 and @xmath73 as determined above .
the data again show considerable curvature but the fit to the assumed functional form is reasonable over the range @xmath89 . finally , in figures ( [ fig6 ] ) and ( [ fig7 ] ) we display the scaled form of @xmath90 , the probability that a gel particle is part of the percolating cluster using the predicted exponent ratios @xmath76 for @xmath46 and @xmath91 for @xmath38
. these two figures present the least impressive collapse of data to a universal curve , especially at the larger values of @xmath92 .
one can improve the collapse by different choice of @xmath93 and @xmath20 but at the expense of violating hyperscaling .
we also note that the data for the two largest values of @xmath45 are reasonably close to each other over the entire range of @xmath94 .
we have also carried out a limited number of simulations for @xmath95 and @xmath96 with the crosslinking process described above . in both cases , the critical exponents and the cluster size distributions are entirely consistent with ordinary three - dimensional percolation .
this suggests that either there is a critical gel fraction @xcite below which the geometric properties of the clusters are described by continuously varying exponents or that the apparent variation of the exponents with @xmath8 described above is a finite - size artefact .
only simulations of larger systems can resolve this issue .
we have calculated the shear viscosity for systems up to size @xmath97 as function of the crosslink density @xmath1 for @xmath46 and for @xmath98 for @xmath38 .
systems are equilibrated as a liquid , crosslinked as described above and then evolved by constant energy md for 40000 or 80000 time steps , depending on the crosslink density . here
we have typically used 500 to 2000 different realizations of the crosslinks at each @xmath1 .
we calculate , as in @xcite , the stress - stress autocorrelation function @xmath99 where @xmath100 are elements of the stress tensor . here
the sum is over both gel and solvent particles and @xmath101 is the derivative of the pair potential between particles @xmath102 and @xmath103 .
the analysis of the stress - stress correlation function has been described in @xcite and is done in the same way here . as @xmath104 , @xmath105 decays extremely slowly and is fitted , at long times , to a stretched exponential .
the static shear viscosity is then obtained from the appropriate green - kubo formula @xcite , @xmath106 the results , for @xmath46 are shown in finite - size scaled form in fig .
( [ fig8 ] ) where @xmath107 is plotted as function of the scaled concentration @xmath94 @xcite .
in contrast to our previous result for @xmath108 and instantaneous crosslinking where we found @xmath6 , we find that @xmath109 provides an excellent collapse of the data with @xmath51 .
we note that , outside the critical region , consistency of the finite - size scaling ansatz requires the scaled viscosity to vary as @xmath110 and it is clear that the data are consistent with this behavior .
we have also calculated the shear viscosity for @xmath38 for @xmath98 .
the raw data are displayed in figure ( [ fig9 ] ) as function of @xmath111 together with the corresponding results for @xmath46 .
fitting to a power law outside the critical region produces an exponent @xmath112 suggesting , as in the case of the static properties , a variation of critical exponents with @xmath8 and an absence of universality .
in this article we have proposed and investigated a new model for gelation which incorporates a solvent on a microscopic level .
for relatively small concentrations of gel , the geometric properties of the system close to the gel point seem to depend continuously on this gel fraction and are , at least for the system sizes investigated , markedly different from three - dimensional percolation .
in particular , the fractal dimension of the clusters seems to be smaller than those of percolation clusters and this more spidery morphology may be responsible for the slower divergence of the shear viscosity as the gel point is approached .
the change in the exponents controlling the geometric properties is rather small and further study of larger systems is certainly necessary to confirm this result .
however , the exponent @xmath4 that characterizes the divergence of the shear viscosity at the gel point is reduced by almost a factor of 2 from its value in the absence of solvent and it is unlikely that this can be attributed to finite - size effects . in light of this result , it seems implausible that a single universality class describes the behavior of transport coefficients and , presumably , the moduli of the amorphous phase near the gel point . the considerable dispersion found in experimental values of the critical exponents @xcite is another indicator that this may be the case . in future work
we intend to explore this new model in greater detail . it will be interesting to investigate if the exponent @xmath4 and the static exponents are tunable by varying the concentration of the solvent and the solubility of the solute .
we also intend to study diffusion constants as function of cluster size and to investigate the existence of long time tails .
finally , one of the original motivations for this model is the existence of a body of experimental work that has yielded values in the range 1.1 1.3 for the viscosity exponent @xmath4 .
clearly , we have moved further from this range of values compared to our previous results .
if the cluster size distribution and cluster geometry is the determining factor in the critical behavior of the transport coefficients then this indicates that models that produce more compact rather than more tenuous clusters than those arising from percolation may be appropriate .
one of us ( bj ) thanks the physics department at simon fraser university for its hospitality during a sabbatical visit .
we thank ralph colby , paul goldbart and sune jespersen for helpful discussions .
this research is supported by the nserc of canada .
see for example d. stauffer and a. aharony , _ introduction to percolation theory _ , 2nd edition , ( taylor and francis , london , 1994 ) .
see d. stauffer , a. coniglio and m.adam , adv .
sci . * 44 * , 103 ( 1982 ) for a review .
d. vernon , m. plischke and b. jos , phys .
rev e * 64 * , 03105 ( 2001 ) .
p.g . de gennes , j. phys ( paris ) , * 40 * , l197 ( 1979 ) . k. broderix , h. lwe , p. mller and a. zippelius , europhys .
lett . , * 48 * , 421 ( 1999 ) ; phys .
e * 63 * , 011510 ( 2001 ) .
m. adam , m. delsanti , d. durand , g. hild and j.p .
munch , pure appl .
, * 53 * , 1489 ( 1981 ) ; m. adam , m. delsanti and d. durand , macromolecules , * 18 * , 2285 ( 1985 ) ; d. durand , m. delsanti and j.m .
luck , europhys .
* 3 * , 297 ( 1987 ) .
lusignan , t.h .
mourey , j.c .
wilson and r.h .
colby , phys .
e * 52 * , 6271 ( 1995 ) ; j.e .
martin and j. wilcoxon , phys .
lett . , * 61 * , 373 ( 1988 ) ; d. adolf and j.e .
martin , macromolecules , * 23 * , 3700 ( 1990 ) ; j.e .
martin , j. wilcoxon and j. odinek , phys . rev . a * 43 * , 858 ( 1991 ) ; j.e .
martin , d. adolf and j. wilcoxon , phys .
lett . , * 61 * , 2620 ( 1988 ) .
powles and d.m .
heyes , mol .
phys . * 98 * , 917 ( 2000 ) .
these authors have studied the properties of systems with a pair potential of the form @xmath113 , for @xmath114 , including our case @xmath115 .
gimel , d. durand and t. nicolai , phys .
b * 51 * , 11348 ( 1995 ) .
a. hasmy and r. jullien , phys .
e * 53 * , 1789 ( 1996 ) .
m. plischke , s. jespersen and b. jos , unpublished .
a critical volume fraction appears in the model of gimel _ et al . _
@xcite , albeit in a rather trivial way : when the volume fraction is increased above 0.31 , the system percolates instantaneously ( in the thermodynamic limit ) and the features associated with cluster - cluster aggregation disappear .
allen and d.j .
tildesley , _ computer simulation of liquids _ ( oxford university press , new york , 1987 ) , chap . 2 ; j.p .
hansen and i.r .
macdonald , _ theory of simple liquids _
( academic press , new york , 1986 ) . a fit of the raw data for @xmath98 and @xmath116 to the form @xmath117 produces the estimates @xmath118 and @xmath119 , in good agreement with the percolation theory estimate of @xmath2 and with the finite - size scaling analysis reported above . for a review ,
see m. adam and d. lairez in _ the physical properties of polymeric gels _ , j.p .
cohen addad , ed .
( john wiley and sons , ltd . , new york , 1996 ) | we study a two - component model for gelation consisting of @xmath0-functional monomers ( the gel ) and inert particles ( the solvent ) .
after equilibration as a simple liquid , the gel particles are gradually crosslinked to each other until the desired number of crosslinks has been attained . at a critical crosslink density the largest gel cluster percolates and an amorphous solid forms .
this percolation process is different from ordinary lattice or continuum percolation of a single species in the sense that the critical exponents are new .
as the crosslink density @xmath1 approaches its critical value @xmath2 , the shear viscosity diverges : @xmath3 with @xmath4 a nonuniversal concentration - dependent exponent . |
understanding the nature of complexity in np - hard optimization problems is an important goal in the study of complex systems @xcite .
most of optimization problems are indeed a constraint satisfaction problem and satisfiability problem is a prototype of such problems @xcite . in a satisfiability problem one looks for a configuration of logical variables that satisfies a number of logical constraints ( clauses ) .
a clause is in turn disjunction of a few logical variables . in random k - satisfiability ( k - sat )
problem a clause involves @xmath0 randomly selected variables each negated with a given probability . increasing the number of clauses per variables , these problems display a phase transition from satisfiable to unsatisfiable phase @xcite .
it is around this sat - unsat transition that the typical complexity of the problem raises rapidly .
in fact the transition is preceded by a clustering of solutions in the configuration space , a phenomenon known as replica symmetry breaking in physics literatures @xcite .
there are a few analytic methods and algorithms , such as replica and cavity methods and survey propagation algorithm @xcite , that deal with the behavior of random satisfiability problems .
+ here we are going to look at the problem from a rather geometrical point of view .
we know that there are efficient algorithms such as walksat and belief propagation @xcite that can efficiently solve a random k - sat problem possessing a tree factor graph ; a factor graph is a bipartite - graph representation of constraint satisfaction problems .
we also know that spanning trees are the closest tree structures to a given graph .
then it is interesting to find the relation between the solutions of a factor graph and the solutions of its spanning trees .
this relation could result in the reduction of the original problem to easier ones .
+ in this paper we show that there is indeed a correspondence between the solutions of a random k - sat problem and those of spanning trees in the associated factor graph . in principle to check that a random k - sat problem is satisfied or not one has to check the satisfiability of all the spanning trees .
the main problem is that in general the number of spanning trees grows exponentially with the size of graph .
moreover we will see that even when we are in the sat phase almost all randomly generated spanning trees are unsatisfiable . to get around this problem
we assign a weight to each edge of the factor graph . for an edge
, its weight is taken as the probability that the edge to be present in the graph .
then we suggest an ansatz that relates the above probability to the survey of warnings sent along the edge and run survey propagation algorithm .
this algorithm converges with a nonzero probability in the sat phase and we find that in this case all maximum spanning trees ( which have the maximum sum of weights on their edges ) are satisfied . in other words ,
the above algorithm defines null edges of the factor graph .
+ the structure of paper is as follows ; in section [ 2 ] we define the problem more precisely and illustrate the relation between a factor graph and its spanning trees .
we briefly deal with the satisfiability of randomly generated spanning trees in section [ 3 ] . in section
[ 4 ] we modify the survey propagation algorithm to generate satisfiable spanning trees . section [ 5 ] includes our concluding remarks .
a random satisfiability problem is defined as follows : we take @xmath1 logical variables @xmath2 .
then we construct a formula @xmath3 of @xmath4 clauses joined to each other by logical and .
each clause contains a number of randomly selected logical variables . in random k - sat problem
each clause has a fixed number of @xmath0 variables .
these variables , which join to each other by logical or , are negated with probability @xmath5 and appear as such with probability @xmath6 .
for example @xmath7 is a 2-sat formula with @xmath8 clauses and @xmath9 logical variables .
a solution of @xmath3 is a configuration of logical variables that satisfy all clauses . + the relevant parameter that determines the satisfiability of @xmath3 is @xmath10 . in the thermodynamic limit ( @xmath11 and @xmath12 ) @xmath3 is satisfied with probability one as long as @xmath13 .
moreover , it has been found that for @xmath14 the problem is in the hard - sat phase @xcite . at @xmath15
we have a dynamical phase transition associated with the break down of replica symmetry .
assuming one - step replica symmetry breaking , one obtains @xmath16 and @xmath17 for random 3-sat problems with @xmath18 @xcite .
although this approximation seems to be exact near the sat - unsat transition but it fails close to the dynamical transition where higher order replica symmetry breaking solutions are to be used @xcite .
the above approximation allows us to compute the configurational entropy or complexity of the problem .
the complexity , which is logarithm of the number of clusters in the solution space , takes abruptly a nonzero value at @xmath15 and decreases to zero at @xmath19 .
+ the factor graph is a bipartite graph of variable nodes and function nodes ( clauses ) .
the structure of this graph is completely determined by an @xmath20 matrix with elements @xmath21 ; @xmath22 if clause @xmath23 contains @xmath24 , it is equal to @xmath25 if @xmath26 appears in @xmath23 and otherwise @xmath27 . + now consider a random k - sat problem and the associated factor graph . in the sat phase the problem has some solutions that satisfy all the clauses .
the factor graph has a number of spanning trees which have their own solutions .
these solutions can be easily obtained by walksat algorithm or some other local search algorithms @xcite .
notice that a spanning tree could have no solution even when the original problem is satisfied .
here we show that ( i ) _ each solution of a spanning tree is the solution of the original graph too . _ and ( ii ) _
each solution of the original graph is the solution of at least one of the spanning trees_. + note that in a spanning tree each function node is connected to a smaller number of variable nodes than in the original graph .
thus if a configuration of variable nodes satisfies a spanning tree it should also satisfy the original problem . + to verify the second statement we start with a solution of the problem and try to construct a spanning tree that is also satisfied by this solution . to this end
we have to remove some of the edges in the graph until no cycle remains in the factor graph . during this process
we should , of course , respect the graph connectedness and more importantly satisfaction of all the function nodes .
therefore , at each step we consider a cycle of the graph and remove one of the null edges along this cycle .
a null edge is an edge that its removal dos not change the number of satisfied clauses .
this removal also preserves the graph connectedness .
note that a null edge at a given stage may not be a null one in the next stages .
+ to complete the proof we should show that for a solution of the factor graph , any cycle in the graph has at least one null edge .
first notice that each function node which lies on a cycle has at least two neighbors .
since we are dealing with a solution then at least one of its neighboring variable nodes satisfies the function node . in this case
at least one of the edges along the cycle will be a null edge and we can freely remove it from the graph .
+ indeed for any solution , a cycle would have a large number of null edges which result in a large degree of freedom in constructing the desired spanning tree .
the number of null edges along a cycle would be of order of the number of function nodes in the cycle .
it means that an arbitrary solution appears in a large number of spanning trees .
+ suppose that we have a satisfiable spanning tree .
then we can use a local search algorithm to find its solutions .
notice that for a tree factor graph there is no frustration in the problem and one can safely assume all the solutions in the same cluster . indeed , warning propagation algorithm @xcite which is based on this assumption gives exact results for tree factor graphs .
thus we expect that the solutions of a satisfiable spanning tree lay in the same cluster of solutions . in the thermodynamic limit ,
it means that for any two solutions , there is at least one path in the solution space that connects them to each other .
the path starts from one of the solutions and reach the other one by successive steps of finite lengths . since the solutions also satisfy the original problem we may conclude that they lay in the same cluster of the original problem too .
+ please notice that one could use the same arguments ( as given in this section ) to show that any configuration of variables that violates @xmath28 clauses in the original factor graph also appears in at least one of the spanning trees .
therefore , the state with minimum @xmath28 ( the ground state ) in the original factor graph , has also the minimum @xmath28 among the ground states of all the spanning trees .
there is an algorithm that uses a simple random walker on the original factor graph and generates ( with a uniform measure ) a spanning tree @xcite . using this algorithm we generated up to @xmath29 spanning trees of the factor graph of a random 3-sat problem with @xmath30 .
belief propagation algorithm can be used to compute the entropy ( logarithm of the number of solutions ) of each spanning tree .
we found that when @xmath31 both the average entropy and its dispersion decrease exponentially with @xmath32 .
however , if we increase @xmath5 slightly , we find that almost all the randomly generated spanning trees have no solution .
for instance , with @xmath33 and @xmath34 , there is almost no satisfiable spanning tree among @xmath29 randomly generated ones . in this case , in average about
@xmath35 clauses are not satisfied . when we increase @xmath5 to @xmath36 the average number of unsatisfied clauses in a randomly generated spanning tree increases to @xmath37 whereas the original problem is satisfiable .
we found that any naive attempt , like the generalization of simple random walk procedure to a biased one , fails in generating satisfiable spanning trees .
these observations show that finding satisfiable spanning trees is not an easy task . in the following
we look for a more sophisticated way to solve this problem
. we will restrict ourselves to random 3-sat problems in the most random case , i.e. @xmath18 .
in section [ 2 ] we used the notion of null edges to construct spanning trees that are satisfied by a given solution .
it suggests that some of the edges can be removed from the factor graph with no effect on the number of satisfied clauses . but how can we distinguish these null edges ?
suppose that we remove some edges and run the survey propagation algorithm @xcite .
if the problem is in the sat phase the algorithm converges with a nonzero probability and for each edge returns a survey of warnings @xmath38 .
the survey @xmath38 is the probability that in an arbitrary cluster of solutions , clause @xmath23 sends a warning to variable @xmath39 .
this warning enforces @xmath39 to take a value which satisfies @xmath23 .
if the problem is in the unsat phase then the algorithm may still converge but it results to a negative value for the complexity . in the following we will work with a version of survey propagation algorithm in which we assign zero probability to contradictory massages .
then the algorithm can end with a set of contradictory surveys . in this case
a variable receives surveys that suggest different values for it and taking any value results in at least one unsatisfied clause .
+ let us define @xmath40 as the probability that edge @xmath41 to be present in the factor graph
. then we run the survey propagation algorithm taking these probabilities into account .
it means that when function node @xmath23 sends a message to its variable node @xmath39 , it gives more weights to the neighbors having a higher probability of being present in the graph .
more precisely , @xmath42 is given by @xmath43,\ ] ] where @xmath44 is the probability that variable @xmath45 can not satisfy clause @xmath23 .
we also denote by @xmath46 the set of variables belong to clause @xmath23 and by @xmath47 the set of clauses that variable @xmath39 contributes in . in survey propagation algorithm @xmath44
is given by @xcite @xmath48 where @xmath49 \prod_{b\in v_a^s(j)}\left(1-w_{b , j}\eta_{b\rightarrow j}\right),\\ \nonumber \pi_{j\rightarrow a}^s=[1-\prod_{b\in v_a^s(j)}\left(1-w_{b , j}\eta_{b\rightarrow j}\right)]\prod_{b\in v_a^u(j)}\left(1-w_{b , j}\eta_{b\rightarrow j}\right).\end{aligned}\ ] ] here @xmath50 denotes to the set of clauses in @xmath51 that variable @xmath45 appears in them as it appears in clause @xmath23 , see fig .
[ f1 ] . is determined by the set of surveys @xmath52 and weights @xmath53 . in this factor
graph squares and circles represent clauses and variables respectively .
a dashed line shows that the negated variable appears in the clause.,width=302 ] the remaining set of clauses are denoted by @xmath54
. we should also give a way to determine @xmath40 .
note that we expect a positive correlation between @xmath40 and @xmath38 .
if the survey is large , the edge is necessary for satisfaction of clause @xmath23 and in the opposite case its presence is not essential .
our suggested ansatz for the weights is @xmath55^{\mu},\ ] ] where @xmath56 controls the number of removed edges .
for each instance of the problem we start by a random initialization for @xmath57 s and @xmath58 s . in each iteration
we select all of the edges in a random way .
when an edge is selected we first update @xmath38 and then @xmath40 according to eqs .
( [ eta0]-[weta ] ) .
we expect that for appropriate values of @xmath59 the above algorithm converges ( without contradictory surveys ) as long as we are in the sat phase . in practice
we define a maximum @xmath60 for the number of iterations @xmath61 .
thus for a given @xmath59 it is possible not to reach a convergence in @xmath62 .
we define @xmath63 as the probability that a randomly generated instance of the problem converges in a number of iterations less than @xmath60 . in our numerical simulations
we will take @xmath64 and the limit of convergence is taken @xmath65 .
note that when @xmath66 we recover the usual survey propagation algorithm on the original factor graph .
in this situation the algorithm always converges and @xmath67 .
increasing @xmath59 means that we are removing some of the edges and so we expect more time to reach a convergence .
thus @xmath68 would be a decreasing function of its argument .
we find that its behavior depends on @xmath32 , fig .
in fact , @xmath63 diminishes more rapidly for greater values of @xmath32 . + ) and satisfiability probability of a maximum spanning tree ( @xmath69 ) versus @xmath59 for a random 3-sat problem with @xmath70 and @xmath18 . the data for @xmath63 obtained from @xmath71 realizations of the problem . to obtain @xmath69 we randomly generated @xmath72 maximum spanning trees in @xmath73 satisfiable realizations . , width=302 ] suppose that our algorithm converges .
then we expect that a spanning tree with maximum sum of weights for its edges , i.e. a maximum spanning tree , be a satisfiable one . when the algorithm converges and returns the @xmath58s we can construct a maximum spanning tree as follows : starting from a randomly selected node in the original factor graph we first find the maximum weight among the edges that connect it to the other nodes
. then we list the edges having a weight in the @xmath74-neighborhood of the maximum one and add randomly one of them to the new factor graph .
if we repeat the addition of edges @xmath75 times we obtain a spanning tree factor graph which is a maximum spanning tree .
notice that taking a nonzero interval to define the edges of maximum weight at each step , along with the randomness in choosing one of them , allow to construct a large number of maximum spanning trees .
let us define @xmath76 as the probability that a maximum spanning tree be satisfiable if the algorithm converges .
the best value of @xmath59 is one that for which @xmath69 takes its maximum value . in fig .
[ f2 ] we also show @xmath76 for two values of @xmath77 . as the figure shows @xmath78 when @xmath79 ( @xmath80 ) and @xmath81 ( @xmath82 ) .
however , as stated above , by increasing @xmath59 the convergence time increases .
thus the optimal @xmath59 for @xmath77 will be around @xmath83 and @xmath84 respectively .
the optimal @xmath59 decreases with @xmath32 .
it means that as we get closer to the sat - unsat transition we have to give more importance to the surveys . + ) and satisfiability probability of a maximum spanning tree ( @xmath69 ) versus @xmath59 for random 3-sat problems of different sizes with @xmath18 . here
the statistical errors are of order @xmath85.,width=302 ] to see the finite size effects we run the algorithm for larger problem sizes @xmath86 .
the results for @xmath63 and @xmath69 at @xmath82 have been shown in fig .
the convergence probability decreases more rapidly with @xmath59 as @xmath1 increases .
that is we have to pay more computational efforts to get a converged situation . on the other side , by increasing @xmath1 the satisfiability probability enhances for smaller values of @xmath59 and gets more rapidly its saturation value .
we hope that this behavior of @xmath69 compensate the decrease in @xmath63 for larger problem sizes . as fig .
[ f3 ] shows the changes in the quantities are significant and we have a strong size dependence for the small values of @xmath1 studied here .
+ finally let us look at the statistics of @xmath40 s and the organization of maximum spanning trees in the space spanned by spanning trees .
+ we observe that if @xmath78 then the weight distribution of the edges , @xmath87 , is almost sum of a few delta peaks .
[ f4 ] displays this behavior of @xmath87 for @xmath77 . in both cases , edges of weight @xmath88
have a considerable contribution and the weight distribution dose not define a single characteristic weight .
we think that it is these properties of @xmath87 that allow the maximum - spanning - tree scenario to work . a clear difference between the two distributions in fig .
[ f4 ] is the presence of a delta peak at @xmath89 for @xmath82 .
we found that this peak appears continuously around @xmath90 , where we expect the dynamical phase transition to happen .
+ , @xmath18 and @xmath91 when @xmath59 is @xmath92 respectively .
the data are result of averaging over @xmath93 realizations , width=302 ] we define the distance between two spanning trees as the number of uncommon edges . then we take two randomly generated satisfiable spanning trees and obtain their distance @xmath94 .
we repeat this for a number of times and construct @xmath95 , the normalized number of times that the two spanning trees are at distance @xmath94 of each other .
the result has been shown in fig .
[ f5 ] for @xmath96 . in these cases @xmath95 is a gaussian distribution and defines a characteristic distance which grows with @xmath32 .
moreover , as the figure displays , by increasing @xmath32 dispersion around the characteristic distance increases . note that more distant factor graphs are more likely to have distant solutions .
therefore , we may conclude that as satisfiable spanning trees get away from each other the same happens in the solution space .
these facts are consistent with the present picture @xcite of the clustering phenomenon in the solution space
. however , the presented data in fig .
[ f5 ] are not sufficient to detect the clustering phase transition in the solution space .
to this end we should , probably , study the behavior of the average distance and its dispersion for different problem sizes . for a single realization of random 3-sat problem with @xmath70 and @xmath18 . for @xmath97 ( @xmath98 respectively )
we have generated @xmath99 pairs of satisfiable spanning trees .
, width=302 ]
we showed that there is a correspondence between the solutions of a random k - sat problem and those of spanning trees in the associated factor graph .
this relation may be useful in finding more rigorous results for the structure of the solution space .
the concept of null edges helped us to modify survey propagation algorithm and to construct satisfiable spanning trees .
a satisfiable spanning tree has a number of solutions ( in the same cluster of solutions ) that can easily be obtained by local search algorithms .
this provides us another way of extracting solutions of a random k - sat problem .
we used the modified algorithm to study organization of satisfiable spanning trees in the space of spanning trees .
we found that by increasing the number of clauses per variable , the distance between satisfiable spanning trees increases .
this finding supports the previous results on the clustering in the solution space .
+ in this paper we focused on the case @xmath100 and @xmath18 but we found that the algorithm works well for @xmath101 and other values of @xmath5 ( not represented here ) . | working with tree graphs is always easier than with loopy ones and spanning trees are the closest tree - like structures to a given graph .
we find a correspondence between the solutions of random k - satisfiability problem and those of spanning trees in the associated factor graph .
we introduce a modified survey propagation algorithm which returns null edges of the factor graph and helps us to find satisfiable spanning trees .
this allows us to study organization of satisfiable spanning trees in the space spanned by spanning trees . |
the small mode volume of evanescent field atom traps engenders strong atom - light interactions without the need for a cavity @xcite .
the recent demonstrations of trapping @xmath3cs with an optical nanofiber ( onf ) @xcite - and a state - insensitive variant @xcite mark an important experimental realization of these systems .
their high optical depth ( @xmath4 ) allows for efficient dispersive readout @xcite and strong nonlinear interactions , and they could potentially realize collective effects such as superradiance @xcite . the success of onf traps has inspired a growing effort to trap atoms in the evanescent field of nanophotonic waveguides @xcite .
this regime of strong coupling opens the door to the study of long - range interactions and the formation of so - called atomic mirrors @xcite , the observation of self - crystallization @xcite , or the generation of a sub - poissonian atom number distribution @xcite .
furthermore , low loss onfs @xcite have been proposed as one of steps toward the realization of a hybrid quantum system coupling photons to atoms to superconducting circuit elements @xcite .
optical dipole trapping of atoms is a well - developed technology applied to numerous atomic species .
the extension of optical trapping to evanescent fields of an onf shares similarities with dipole trapping with free space beams , but has one distinction - the evanescent field may have a substantial longitudinally polarized component of the electric field .
this can lead to surprisingly large differences in the absorption of probe light for two different species , even when they are both alkali atoms ( e.g. rb and cs ) , due to the effects of the vector light shift .
our system traps atoms with two lasers , achieving trap depths of a few hundreds of microkelvin .
we can not simply determine atom number by the absorption of a probe beam by an optically thick medium with a lorentzian line shape .
distinct asymmetries are observed that we trace to the effects of the vector light shifts associated with the optical trapping fields , and their inherent elliptical polarization with an appreciable component along the direction of propagation .
although rb and cs are nominally atoms with very similar atomic structure , the light shifts can in fact be quite different , with differential light shifts much larger in rb than cs , leading to a modified absorption profile .
the paper is organized as follows .
[ setup ] outlines the experimental setup .
we present experimental confirmation of our trap in sec .
we introduce a theoretical model based on light shifts , finite atom temperature , and population redistribution in sec .
[ theory ] , and use it to study the inhomogeneous absorption profile .
[ conc ] summarizes our findings and provides an experimental outlook .
our system consists of the main science chamber ( msc ) with the antechamber ( ac ) equipped with a precision vacuum manipulator ( vm ) as shown in fig .
[ fig_setup ] .
we transfer onfs from the antechamber to the science chamber without breaking the science chamber vacuum .
the science chamber maintains a pressure of @xmath5 mbar with two ion pumps ( @xmath6 ) .
we produce the onf by thinning a single - mode fiber ( fibercore ) in an in - lab setup @xcite .
the onfs typically have greater than 99% transmission for waists with diameters of 500 nm and lengths from 1 mm to 10 cm .
the data presented in this paper are taken using an onf with a waist of 530 nm diameter and 7 mm length , with a tapering region of 39 mm in length .
the produced onf epoxied onto the titanium alloy fiber holder ( fh ) is held on the vacuum manipulator rod ( vmr ) at the antechamber and is transfered to the science chamber .
the mounted onf is extended to the out - of - vacuum patch - cord single - mode fiber via the teflon ferrule fiber feedthrough ( ff ) with a swagelok .
a magneto - optical trap ( mot ) loaded from a background vapor of @xmath0 produces a cloud of @xmath7 atoms .
we overlap the cloud with the onf waist using magnetic field shim coils and the vacuum manipulator ( vg scienta transax ) with 2-d manual translation stages ( ts ) .
two orthogonal imaging systems ensure alignment .
atoms are loaded into the onf trap ( left on throughout the experiment ) after 90ms of increased mot detuning and a 1ms duration optical molasses stage .
the sub - doppler cooling during this loading stage yields mot temperatures of @xmath8 , as determined by time - of - flight measurements .
an onf trap requires light tuned red of resonance ( with respect to to the @xmath9rb d2 line ) to provide an attractive potential and light tuned blue of resonance to prevent atoms from striking the onf surface .
a 750-nm wavelength laser ( coherent ti : sa 899 ) provides the repulsive force , and a 1064-nm wavelength beam ( jdsu npro ) in a standing wave configuration ( to provide longitudinal confinement ) provides the attractive potential .
a potential minimum of 370@xmath10 in depth is formed @xmath11215 nm from the fiber surface , and the trapping frequencies are calculated to be ( @xmath12 , @xmath13 , @xmath14 ) = ( 253 , 371 , 104)khz .
all of the onf trapping beams and the near - resonant probe beam are intensity - stabilized .
the amplitude lock of the probe beam includes a sample - and - hold to stabilize the power in between pulses so that the servo can rapidly recapture when the pulse is turned on , minimizing transients .
we measure atomic absorption with a weak , near - resonance beam ( 780 nm ) coupled through the onf , counting transmitted photons with avalanche photodiodes ( apd , laser components count-250c - fc ) operating in geiger mode .
because light levels near 10pw saturate the apds , great care must be taken to filter stray light and maintain low probe power .
three narrow - line volume bragg gratings ( vbg , optigrate bp-785 , 0.18 nm spectral bandwidth at 785 nm ) filter amplified spontaneous emission from the ti : sapphire laser near 780 nm . a vbg at the output of the nanofiber serves as a mirror to direct signal to the apds and as another filter to block in - fiber background induced by the blue trapping beam .
this light due to either fluorescence or raman scattering is the main source of background in the experiment .
two more bandpass filters further reduce background counts , and finally long - pass color filters ( thorlabs , fgl645 ) directly in front of the apd fiber couplers reduce short - wavelength background from stray light .
a series of differing optical depth neutral density filters before and after the nanofiber allow us to vary the probe intensity while keeping light levels within the dynamic range of the apds .
ttl pulses from the apds are counted with a field - programmable gate array ( fpga ) and processed to extract absorption signals and full photon counting statistics .
rb atoms , the standing waves with nd : yag laser ( 1064 nm ) and the travelling wave with ti : sapphire laser ( 750 nm ) create a nanofiber atom trap .
an absorption signal is measured by a sub - pw 780 nm probe .
see text for details .
ac : antechamber ; aom : acousto - optic modulator ; apd : avalanche photodiode ; bpf : band - pass filter ; dm : dichroic mirror ; fh : fiber holder ; ff : fiber feedthrough ; gt : glan - thompson polarizer ; gv : gate valve ; ip : ion pump ; @xmath15 : half - wave plate ; mc : mot coil ; mot : magneto - optical trap ; msc : main science chamber ; oi : optical isolator ; pbs : polarizing beam spliter ; pd : photo diode ; pid : proportional - integral - derivative lock box ; ts : 2-d translation stage ; vbg : volume bragg grating ; vm : vacuum manipulator ; vmr : vacuum manipulator rod . ]
the absorption profiles are measured via an in - fiber analog of standard absorption spectroscopy .
we use two probe pulses ; the first pulse measures the atomic absorption signal ( @xmath16 ) , and the second pulse is a reference signal with no atoms ( @xmath17 ) . in between the two probe pulses , the 1064 nm trapping beam is turned off and a slightly blue - detuned laser from the mot beam paths kicks away the trapped atoms . based on the ratio of these probe signals
, we calculate the measured transmission @xmath18 , where @xmath19 is the background apd signal with no probe light , with contributions from detector dark counts and fiber - induced fluorescence . for a single lorentzian lineshape with width @xmath20 , one can easily estimate the optical depth @xmath4 by fitting @xmath21 $ ] to the data .
the total number @xmath22 of trapped atoms is then given by @xmath23 , where @xmath24 is the single - atom optical depth .
we calculate @xmath25 to be @xmath26 by comparing the atomic absorption cross section to the optical nanofiber mode area .
our measured absorption profile displays a markedly asymmetric lineshape ( see fig . [ fig_absorption ] ) .
therefore , it is not trivial to estimate the number of trapped atoms . for a given number of trapped atoms , any broadening mechanisms serve to reduce the maximum absorption .
in addition , the maximum absorption point sets a lower limit on the @xmath4 and the number of trapped atoms , i.e. @xmath27 $ ]
. we will discuss this more in sec .
[ theory ] , where we develop a method to estimate the number of trapped atoms based on these asymmetric absorption profiles . statistical error ; red - circle : averaged absorption signal with 50shots ) .
inset illustrates a typical trap lifetime measurement with a fit to a decaying exponential ( @xmath28 = 23ms ) .
the absorption measurement is done with red - detuned ( 3.35mw@xmath292 , standing waves ) and blue - detuned ( 7.4mw , travelling wave ) trapping beams .
the trap configuration is close to the pp case with an angle ( 23.5@xmath30 ) . ]
we observe trapping lifetimes ( without any additional cooling ) of approximately 23ms ( see fig .
[ fig_absorption ] ( inset ) ) .
we observe that our onf exhibits large - amplitude transverse vibrations near 550hz .
these are too low in frequency to produce heating as it would be adiabatic motion in terms of the optical trapping potential .
though this is far away from any relevant trap frequency , the acceleration of the fiber may be high enough at times to affect loading , as the macroscopic motion of the trap is no longer adiabatic relative to the mean atomic motion .
this lifetime is typical of onf traps , which are generally shorter - lived than standard optical dipole traps , for reasons that are not yet fully understood .
different polarization configurations of the red- and blue - detuned trapping beams can provide trapping , including parallel polarization ( pp ) and cross polarization ( cp ) .
the pp configuration requires less blue - detuned light than the cp configuration , but has larger vector light shifts .
the cp configuration results in more azimuthally localized potentials and smaller vector light shifts . in this paper , we chose a trapping geometry close to the pp configuration with an angle ( 23.5@xmath30 ) between the polarizations of red- and blue - detuned trapping beams where we measured the deepest optical depth ( @xmath1198% absorption ) , better than with the pp or cp configurations ( 50@xmath1170% absorption ) .
we analyzed the polarizations of nanofiber modes with rayleigh scattering , confirming the local polarization angles , but do not have a reason why this particular polarization angle produced the best optical depth .
2 ) and blue - detuned ( 7.4mw ) trapping beams .
the trap configuration is close to the pp case with an angle ( 23.5@xmath30 ) . ( a )
light shifts with linearly polarized trapping beams of @xmath31 s @xmath32 state and @xmath33 s @xmath34 state .
zeeman sub - levels of @xmath33 are degenerate because of no vector and tensor light shifts , and zeeman sub - levels of @xmath31 are split due to tensor light shifts .
( b ) light shifts with circularly polarized trapping beams of @xmath31 s @xmath32 state and @xmath33 s @xmath34 state .
inset shows the cross - sectional view of light shifts . in fig .
[ fig_light_shift ] , the color lines of red , red - dash , purple , black , cyan , blue - dash , and blue correspond to @xmath31 s @xmath35 state , respectively , and the lines of red , purple , black , cyan , and blue correspond to @xmath33 s @xmath36 state , respectively .
in addition , the up - down arrow means the light - shifted optical transition frequency at the trapping position . here , the quantization axis is defined along the propagation direction ( @xmath37 ) of the probe beam ( see fig . [ fig_setup ] ) , and the light polarization is defined in the @xmath38 plane with stokes vectors ; the linearly ( circularly ) polarized light means @xmath39 = 0 ( @xmath40 ) . ] to calculate the inhomogeneous broadening of the absorption line , we need to include the light - shifted optical transition frequency for atoms trapped in the optical fiber potential .
this requires appropriate weighing over the polarization of the modes , the @xmath41-state distribution of the atoms , and the location of the atoms within the trap due to thermal motion .
qualitatively , lower temperature atoms moves the absorption profile to the blue as the atoms spend more time in a deeper trap , which leads to more light shifts , and higher temperature atoms determine the left - wing of the profile by the boltzmann truncation .
the light - shifted optical transition frequencies of @xmath0 atoms at trapping positions move the absorption profile to the blue , and the vector light shifts broaden the blue - side of the profile .
the vector light shift is related to the ellipticity of the light and the populated atomic states .
when the light propagation direction and the populated atomic state @xmath42 are collinear along the quantization axis ( @xmath37 ) , the ellipticity is defined in the @xmath38 plane with stokes vector components @xmath43 @xcite , and the vector light shift ( @xmath44 ) is proportional to the ellipticity of the light , where @xmath45 is calculated from wigner - eckart theorem . the linearly polarized light ( @xmath46 ) can be symmetrically decomposed into right - hand and left - hand circularly polarized lights ( @xmath47 , respectively ) that cancel the ellipticity . the @xmath48 fundamental mode has an axial - direction ( @xmath37 ) electric field component that leads to ellipticity ( @xmath49 ) in the @xmath38 plane , which creates the vector light shifts ( see fig .
[ fig_light_shift ] ( b ) ) . instead of considering all the ellipticities for each trapping location
, we add linearly and circularly polarized light shifts with a variable ratio ( see fig . [ fig_fitting ] ) .
a complete description of the atomic absorption has contributions from homogeneous ( natural linewidth ) broadening @xmath50 and inhomogeneous broadening @xmath51 , generally resulting in the symmetric voight profile @xcite as follows : @xmath52 where @xmath53 is the convolution of lorentzian and gaussian profiles . for the optical transition of @xmath9rb atoms
, we study the inhomogeneous broadening to consider the effects of the atomic temperature distribution , zeeman - sublevel - dependent population distribution , and the light shifts of the hyperfine ground state ( @xmath33 to @xmath54 and @xmath31 ) and the optical excited state ( @xmath31 to @xmath55 , @xmath56 , @xmath57 ) @xcite .
this requires considering scalar , vector , and tensor light shifts @xcite
. the vector light shift can be large due to a non - negligible axial - direction electric field component ( @xmath58 ) in the fundamental mode @xmath48
figure [ fig_light_shift ] shows the light shifts of the hyperfine ground state and the optical excited state , considering the polarization of the blue- and red - detuned trapping beams .
the up - down arrow represents the light - shifted optical transition frequency at the trapping position , which affects the absorption profile .
the linearly polarized trapping beams have no vector light shifts of @xmath33 and @xmath31 ( fig .
[ fig_light_shift ] ( a ) ) . however , the nanofiber s @xmath48 mode always has the propagation direction components related to the vector light shift ( fig .
[ fig_light_shift ] ( b ) ) , and moving atoms in the trap with a finite temperature experiences the variation of light shifts . in the experiment ,
the quantization axis is not well - defined with a residual magnetic field after optical molasses cooling , and the initial states of trapped atoms are assumed to be unpolarized .
the pp trap configuration with an angle induces more vector light shifts from the blue - detuned traveling wave than the red - detuned standing waves . in the case of the absorption profile of the unpolarized atoms requiring the sum of all the zeeman sub - levels absorptions ,
@xmath0 atoms are expected to experience more asymmetric profile than @xmath59 atoms due to the vector light shifts from the upper state manifolds .
because of differences in the atomic structure of higher lying states in rb compared to cs , there is no available magic wavelength and excited state vector light shifts are considerably larger . for a ground state @xmath60 and an excited state @xmath61 represented by @xmath62 and @xmath63 ,
the inhomogeneous term @xmath64 can be defined for trapped atoms having a temperature @xmath65 as follows : @xmath66 where @xmath67 ; @xmath68 is the trapping potential of hyperfine ground states ( @xmath33 ) ; and @xmath69 is the light - shifted optical transition frequency ( @xmath33 to @xmath31 transition ) . @xmath68 and @xmath69 , dependent on powers and polarizations of the two trapping beams , have spatial dependence and need to be integrated over the effective volume of a trap site .
the atoms with a temperature @xmath65 higher than a local trap potential @xmath70 are truncated in the calculation . ) and the linearly polarized light ( @xmath71 ) ; the quantization axis is defined along the propagation direction of the probe beam . for the fittings , we assume red - detuned ( 3.35mw@xmath292 ) and blue - detuned ( 7.4mw ) trapping beams and the trap configuration close to the pp case with an angle ( 23.5@xmath30 ) . ] here , we define a homogeneous profile including light shift broadening as follows : @xmath72 where @xmath73 is the broadened width of an optical transition @xmath74 ; the standard deviations of state - dependent light - shifted optical transitions at each location @xmath75 are calculated for a frequency @xmath74 and averaged over all @xmath76 .
we define transmission @xmath77 ( sec .
[ exp ] ) based on our definition of the inhomogeneous broadening of the optical transition as follows : @xmath78,\\ & \approx & \exp[- n \cdot od_{1 } \int n_{ij}(\omega ' ) \mathcal{l}(\omega-\omega ' ) d\omega'],\end{aligned}\ ] ] where @xmath79 and @xmath80 are defined in eqn .
( [ eqn2 ] , [ eqn3 ] ) , and @xmath81 is the population of zeeman sub - levels determined by optical bloch equations during optical pumping from the probe beam .
@xmath82 is the relative strength of the atomic dipole moment related to the polarization state @xmath83 of the probe , and the od per atom is @xmath84 ( @xmath85 : the effective mode area of the nanofiber probe , @xmath86 : atomic scattering cross - section ) .
this can be regarded as a constant for a given @xmath87 , and assumes no light shifts from the low intensity probe .
given the uncertainties in the exact polarization profile of the optical modes where the atoms are trapped , the @xmath41-state distribution of the atoms , and the degree to which a truncated boltzmann distribution is a correct assumption , we qualitatively use the asymmetric profiles to estimate trapped atom number ( see fig . [ fig_fitting ] ) .
lower @xmath65 moves the profile to the blue as the atoms stay more in higher intensity regions
. a larger fraction of circular polarization broadens the blue - side of the profile by the vector light shifts .
our calculated profiles do not reproduce the feature of less absorption around 15mhz .
this may be related to probe - induced heating near resonance or probe - induced optical pumping that modifies the sublevel populations and thus absorption cross section .
we focus on the left and right tails of the absorption profile to estimate atom number @xmath88 for @xmath65 = 55@xmath89k ; this corresponds to @xmath4 = 8.4 with our calculated @xmath24 = 0.0278 .
based on our highest absorption of 96.8% ( @xmath4 = 3.44 ) at a probe - detuning of 10mhz , the absolute lower bound of trapped atom number is @xmath22 = 123 ( as all broadening mechanisms will decrease the maximum observed absorption and increase the extracted number of atoms ) .
we realized an optical nanofiber atom trap for @xmath9rb using the evanescent fields of 750 nm and 1064 nm beams , and find absorption profiles with much less symmetry than have been observed with cs atoms .
we qualitatively explain the asymmetric broadening behavior as arising from larger differential light shifts arising from the @xmath31 to upper transitions , which are larger in rb than cs . using a model of the broadening
, we estimate the number of trapped atoms .
+ 99 f. l. kien , s. dutta gupta , v. i. balykin , and k. hakuta , `` spontaneous emission of a cesium atom near a nanofiber : efficient coupling of light to guided modes , '' phys .
a * 72 * , 032509 ( 2005 ) .
e. vetsch , d. reitz , g. sagu , r. schmidt , s. t. dawkins , and a. rauschenbeutel , `` optical interface created by laser - cooled atoms trapped in the evanescent field surrounding an optical nanofiber , '' phys .
lett . * 104 * , 203603 ( 2010 ) . j .-
bguin , e. m. bookjans , s. l. christensen , h. l. srensen , j. h. mller , e. s. polzik , and j. appel , `` generation and detection of a sub - poissonian atom number distribution in a one - dimensional optical lattice , '' phys .
* 113 * , 263603 ( 2014 ) .
a. goban , k. s. choi , d. j. alton , d. ding , c. lacrote , m. pototschnig , t. thiele , n. p. stern , h. j. kimble , `` demonstration of a state - insensitive , compensated nanofiber trap , '' phys .
* 109 * , 033603 ( 2012 ) .
s. t. dawkins , r. mitsch , d. reitz , e. vetsch , and a. rauschenbeutel , phys .
lett . * 107 * , 243601 ( 2011 ) .
f. l. kien and k. hakuta , `` cooperative enhancement of channeling of emission from atoms into a nanofiber , '' phys .
a * 77 * 013801 ( 2008 ) .
f. l. kien and k. hakuta , `` slowing down of a guided light field along a nanofiber in a cold atomic gas , '' phys .
a * 79 * 013818 ( 2009 ) .
. hung , s. m. meenehan , d. e. chang , o. painter , and h. j. kimble , `` trapped atoms in one - dimensional photonic crystals , '' new j. phys . *
15 * , 083026 ( 2013 ) .
j. lee , d. h. park , s. mittal , m. dagenais , and s. l. rolston , `` integrated optical dipole trap for cold neutral atoms with an optical waveguide coupler , '' new j. phys .
* 15 * 043010 ( 2013 ) .
d. e. chang , l. jiang , a. v. gorshkov , and h. j. kimble , `` cavity qed with atomic mirrors , '' new .
* 14 * , 063003 ( 2012 ) .
d. e. chang , j. i. cirac , and h. j. kimble , `` self - organization of atoms along a nanophotonic waveguide , '' phys .
* 110 * 113606 ( 2013 ) .
t. grieer and h. ritsch , `` light - induced crystallization of cold atoms in a 1d optical trap , '' phys .
lett * 111 * , 055702 ( 2013 ) .
j. m. ward , a. maimaiti , vu h. le , s. nic chormaic , `` contributed review : optical micro- and nanofiber pulling rig , '' rev sci instrum .
* 85 * , 111501 ( 2014 ) .
j. e. hoffman , s. ravets , j. a. grover , p. solano , p. kordell , j. d. wong - campos , l. a. orozco , and s. l. rolston , `` ultrahigh transmission optical nanofibers , '' aip advances , * 4 * , 067124 ( 2014 ) .
r. nagai and t. aoki , `` ultra - low - loss tapered optical fibers with minimal lengths , '' optics express vol . *
22 * , issue 23 , pp .
28427 - 28436 ( 2014 )
. j. verd , h. zoubi , ch .
koller , j. majer , h. ritsch , and j. schmiedmayer , strong magnetic coupling of an ultracold gas to a superconducting waveguide cavity , " phys
lett . * 103 * , 043603 ( 2009 ) . j. e. hoffman , j. a. grover , z. kim , a. k. wood , j. r. anderson , a. j. dragt , m. hafezi , c. j. lobb , l. a. orozco , s. l. rolston , j. m. taylor , c. p. vlahacos , and f. c. wellstood , `` atoms talking to squids , '' rev .
s * 57 * ( 3 ) , 1 ( 2011 ) , also at http://arxiv.org/abs/1108.4153 .
z. kim , c. p. vlahacos , j. e. hoffman , j. a. grover , k. d. voigt , b. k. cooper , c. j. ballard , b. s. palmer , m. hafezi , j. m. taylor , j. r. anderson , a. j. dragt , c. j. lobb , l. a. orozco , s. l. rolston , and f. c. wellstood , `` thin - film superconducting resonator tunable to the ground - state hyperfine splitting of @xmath9rb , '' aip advances * 1 * , 042107 ( 2011 ) .
m. hafezi , z. kim , s. l. rolston , l. a. orozco , b. lev , j. taylor , `` an atomic interface between microwave and optical photons , '' phys .
a. * 85 * , 020302 ( r ) ( 2012 )
. s. l. rolston , j. r. anderson , u. chukwu , j. a. grover , j. b. hertzberg , j. e. hoffman , p. kordell , j. lee , c. j. lobb , l. a. orozco , s. ravets , p. solano , k. d. voigt , f. c. wellstood , j. d. wong - campos , g. beadie , f. k. fatemi , `` a hybrid quantum system of atoms trapped on ultrathin optical fibers coupled to superconductors , '' proc .
spie * 8875 * , quantum communications and quantum imaging xi , 88750l ( 2013 ) .
j. lee , j. a. grover , l. a. orozco , and s. l. rolston , `` sub - doppler cooling of neutral atoms in a grating magneto - optical trap , '' josa b , vol .
* 30 * , issue 11 , pp . 2869 - 2874 ( 2013 ) .
k. d. voigt , j. b. hertzberg , j. e. hoffman , j. a. grover , p. solano , r. p. budoyo , c. ballard , j. lee , j. r. anderson , c. j. lobb , l. a. orozco , s. l. rolston , f. c. wellstood , `` movable thin - film superconducting resonator coupled to a tapered optical microfiber at 15 mk , '' applied superconductivity , ieee transactions on * 25 * ( 3 ) ( 2015 ) .
b. arora , m. s. safronova , and c. w. clark , `` magic wavelengths for the npns transitions in alkali - metal atoms , '' phys .
a * 76 * , 052509 ( 2007 ) .
j. m. geremia , j. k. stockton , and h. mabuchi , `` tensor polarizability and dispersive quantum measurement of multilevel atoms , '' phys .
a * 73 * , 042112 ( 2006 ) .
i. h. deutsch , p. s. jessen , `` quantum control and measurement of atomic spins in polarization spectroscopy , '' optics communications * 283 * pp .
681694 ( 2010 ) .
f. l. kien , p. schneeweiss , and a. rauschenbeutel , `` state - dependent potentials in a nanofiber - based two - color trap for cold atoms , '' phys .
a * 88 * , 033840 ( 2013 ) .
j. d. jackson , `` classical electrodynamics , '' 3rd ed ( wiley , new york ) ( 1998 ) .
w. demtrder , `` laser spectroscopy , '' springer , third edition ( 2003 ) . | we experimentally demonstrate optical trapping of @xmath0 atoms using a two - color evanescent field around an optical nanofiber . in our trapping geometry , a blue - detuned traveling wave
whose polarization is nearly parallel to the polarization of a red - detuned standing wave produces significant vector light shifts that lead to broadening of the absorption profile of a near - resonant beam at the trapping site .
a model that includes scalar , vector , and tensor light shifts of the probe transition @xmath1-@xmath2 from the trapping beams , weighted by the temperature - dependent position of the atoms in the trap , qualitatively describes the observed asymmetric profile and explains differences with previous experiments that used cs atoms .
the model provides a consistent way to extract the number of atoms in the trap . |
in the current hierarchical clustering models , galaxies form by gas cooling inside virialized cold dark matter haloes in a way that makes their properties depend only on the halo mass in which they form .
recently , results from @xmath12-body simulations have shown that the halo clustering at large - scales depends on the halo formation history in addition to the halo mass @xcite .
this effect , that is not expected from the excursion set theory , was termed `` assembly bias '' .
for example , gao et al .
( 2005 ) found that low - mass haloes which were assembled at high redshifts are more strongly correlated than those of the same mass that assembled more recently .
therefore , one could ask whether the assembly bias could also affect galaxy properties . by means of a semi - analytic model of galaxy formation ,
croton et al .
( 2007 ) showed that galaxy clustering on large scales is affected by the correlation of halo environment with halo assembly history .
lacerna et al .
( 2011 , hereafter paper i ) found an assembly bias effect in the semi - analytic galaxies from the lagos , cora @xmath13 padilla ( 2008 ) model , where the old population has a higher clustering at large scales than the young population with the same host halo mass . on the observational side ,
the existence of the assembly bias is still debated .
for example , skibba et al . (
2006 ) used galaxies from the sloan digital sky survey ( sdss ; york et al .
2000 ) to measure marked correlation functions which in addition to the halo model suggest that the correlation between halo formation and environment is not important for the bright galaxy population .
tinker , wetzel @xmath13 conroy ( 2011 ) found similar results , indicating that this dependence arises just from the correlation between halo mass function and environment .
however , yang , mo , & van den bosch ( 2006 ) and wang et al .
( 2008 ) found that groups , selected from the two degree field galaxy redshift survey (; colless et al .
2001 ) and sdss respectively , with red central galaxies are more strongly clustered than groups of the same mass but with blue central galaxies , this effect being much more important for less massive groups .
zapata et al .
( 2009 ) found that sdss galaxy groups of similar mass and different assembly histories show differences in their fraction of red galaxies .
furthermore , cooper et al . (
2010 ) studied the relationship between the local environment and properties of galaxies in the red sequence . after removing the dependence of the average overdensity on colour and stellar mass by setting a density parameter as a estimator of environment such
as it is independent of these galaxy properties , they still found that galaxies with older stellar populations occupy regions of higher overdensities compared to younger galaxies of similar colours or stellar masses .
also , alonso et al .
( 2012 ) show that galaxies in groups in low global densities are prone to show higher star formation compared to galaxies of equal properties and similar host groups but in high density environments .
these results show that the concept of assembly bias is applicable to the galaxies in addition to dark matter ( dm ) haloes and would then influence the physics of galaxy formation . also , this effect is important when galaxies are used to constrain cosmological parameters .
wu et al . (
2008 ) showed that upcoming photometric catalogues such as the dark energy survey ( des ) and the large synoptic survey telescope ( lsst ; lsst science collaborations et al .
2009 ) can infer significantly biased cosmological parameters from the observed clustering amplitude of galaxy clusters if the assembly bias is not taken into account .
paper i presented a new peak height definition different from the halo virial mass , where semi - analytic galaxies in haloes of a given mass but different age do not show the assembly bias effect in the two - halo regime .
this new definition simply adds mass to that of the dark matter halo in a way that makes old , low - mass haloes suffer the most extreme peak height change with respect to what results of using the virial mass as its proxy . in this paper
we will test this model using galaxies selected from the sdss catalogue with the difference that , instead of using the total mass inside spheres of different radii , we will use the total luminosity inside cylinders along the position of each galaxy , which we will use to make a more adequate proxy for a peak height , in a similar way to what was proposed in paper i. we will discuss the implications of using discrete tracers of the density field that consider neighbour galaxies instead of a smooth density field .
the outline of this paper is as follows . in section [ sec_data ]
we present the catalogues of galaxies from the sdss and also from a semi - analytic model .
we characterize the dependence of age on local density in section [ sec_local_density ] .
we study the assembly bias effect , namely , the large - scale clustering of galaxies of a given luminosity depends on the age , for both the sdss and semi - analytic galaxies in section [ section_two - point_sdss ] .
we adopt a modified version of the model from paper i in section [ sec_prm ] , but using observational features .
the nature of the objects that are being considered with the proxy of the overdensity peak height is shown in section [ sec_peak_lum ] , where we discuss the implications of using discrete tracers of the density field that consider neighbour galaxies instead of a smooth density field .
the conclusions of our results are presented in section [ conclusiones ] .
we use galaxies from the spectroscopic sdss data release 4 ( dr4 , adelman - mccarthy et al . 2006 ) , which contains @xmath14480,000 galaxies .
after cleaning of repeated galaxies , saturated objects , galaxies close to the edges of the survey , etc .
, the sample includes @xmath14400,000 galaxies with measured spectra and photometry in the bands _ u , g , r , i , z _
@xcite that are @xmath15-corrected to redshift 0.1 using ` kcorrect v4_2 ` @xcite .
the main motivation to use these data is the measured luminosity - weighted ages by gallazzi et al .
( 2005 ) , which are essential to this work .
there are 183,482 galaxies with estimates of stellar age in the magnitude range @xmath16 ( table [ tabla_dr4_centres ] ) .
we then select four samples of centres and tracers detailed in tables [ tabla_dr4_centres ] and [ tabla_dr4_tracers ] , respectively .
figure [ mag_z ] shows the absolute magnitude in the @xmath17-band , @xmath5 , as a function of redshift with the choice of volume - limited samples , which avoid observational bias . [
cols="^,^,^,^,^",options="header " , ] [ tabla_dr4_tracers ] = 8.9 cm the sdss galaxies will be useful to study whether the assembly bias effect is present in these observed objects thus affecting their properties , as will be shown in section [ section_two - point_sdss ] .
as was the case in paper i , one of the most important parameters throughout this work is the age . for sdss galaxies ,
we use the luminosity - weighted age ( or , simply , stellar age ) and the @xmath4 index ( balogh et al . 1999 ) .
the latter is defined as @xmath19 + where @xmath20 is the average flux density in the specified bands ( in the rest frame ) , and it is smaller for young stellar populations than for old , metal - rich galaxies . in our samples
there are 183,482 galaxies with estimates of stellar age ( table [ tabla_dr4_centres ] ) and 205,076 with @xmath4 , both in the magnitude range @xmath16 . to study the assembly bias in galaxies , where old objects may have a different clustering than young objects of the same mass or luminosity , it is not convenient to work directly with the stellar age , for example , because this quantity correlates with the mass or the luminosity of galaxies .
for instance , luminous galaxies have , on average , older stellar populations ( see figure [ stage_mr ] ) as indicated by their luminosity - weighted age .
= 8.9 cm in paper i , we presented a definition of age which is independent of the mass or the luminosity through the @xmath18 dimensionless parameter @xmath21 + where for the @xmath22 th galaxy , @xmath23 is its stellar age , @xmath24 is the median stellar age as a function of @xmath5 , and @xmath25 is the dispersion around the median obtained from the 10 and 90 percentiles of the distribution ( error bars in fig .
[ stage_mr ] ) .
then , objects with positive ( negative ) values of @xmath18 correspond to old ( young ) populations with respect to the median distribution of stellar ages of their luminosity .
the scatter in the observational stellar age can be higher than the stellar age obtained in simulated galaxies such as the model described in section [ section_sam ] .
we also use @xmath4 in equation ( [ eq_delta ] ) as a proxy for age considering its dependence on the absolute magnitude @xmath5 ( similar to figure [ stage_mr ] ) .
= 7.6 cm = 7.6 cm we use synthetic galaxies from the sag2 model by lagos , cora @xmath13 padilla ( 2008 , see also lagos , padilla & cora 2009 ) , who combine a cosmological @xmath12-body simulation of the concordance @xmath26cdm universe and a semi - analytic model of galaxy formation to follow the evolution of the barionic component of haloes accross cosmic history .
the numerical simulation consists in a periodic box of 60 @xmath27 mpc on a side that contains 256@xmath28 dark matter particles with a mass resolution of @xmath1410@xmath29 @xmath27 m@xmath30 .
the cosmology used is @xmath31 = 1 , @xmath32 = 0.28 ( with a baryon fraction of 0.16 ) , @xmath33 = 0.72 , @xmath34 = 0.9 , and spectral index @xmath35 = 1 .
dark matter haloes are identified as self - bound structures that contain at least 10 particles using a friends - of - friends ( fof ) algorithm ( davis et al .
another algorithm ( subfind ; springel et al .
2001 ) is applied to these groups in order to find substructures with at least 10 particles .
typically , the most massive subhalo is considered the main subhalo , while the other subhaloes are satellites within a fof halo .
the galaxy population in the semi - analytic model is generated using the merger histories of dark matter haloes and their embedded subhaloes .
the virial mass for galaxies corresponds to the virial mass of their respective dm substructures . for the simulated galaxies
, we will use the mass - weighted stellar age in equation ( [ eq_delta ] ) .
magnitudes in the sag2 semi - analytic model are in the johnson system .
therefore , we apply a correction to transform these magnitudes to the @xmath36 system to compare with sdss galaxies using the equations of jester et al .
( 2005 ) where , for objects with r@xmath37i @xmath38 and u@xmath37b @xmath39 , @xmath40 + and , for objects with just r@xmath37i @xmath38 , @xmath41 + figure [ scatter_mass_mr ] shows scatter plots of halo dm mass ( virial mass as explained above and friends - of - friends in the left - hand and right - hand panels , respectively ) vs. @xmath5 .
both definitions show a strong correlation with the inferred @xmath17-band absolute magnitude , which is strong for the virial mass , thus we are able to use @xmath5 as a proxy of mass , specially that of the substructure @xcite .
one may think that old galaxies have larger mass - to - light ratios than young galaxies of equal luminosity .
table [ tabla_mean_mvir_mr ] shows the mean virial mass for old and young galaxies of equal luminosities using the sag2 semi - analytic model and the semi - analytic model of de lucia & blaizot ( 2007 ) .
therefore , there is no a strong dependence regarding the virial mass when splitting the old and young galaxies for a given luminosity .
furthermore , in only one out of the four samples of our model , the halo mass is lower for the older population , which goes in the opposite direction to the common sense interpretation of the cuts in age .
this is due to the fact that we are looking at narrow ranges in luminosity .
c c c c c + @xmath5 & ages & log@xmath42 & log@xmath43 + @xmath44 @xmath45 @xmath5 @xmath45 @xmath46 & old & 10.6 & 10.7 + & young & 10.8 & 10.8 + + @xmath47 @xmath45 @xmath5 @xmath45 @xmath44 & old & 11.1 & 11.0 + & young & 11.1 & 11.1 + + @xmath48 @xmath45 @xmath5 @xmath45 @xmath47 & old & 11.6 & 11.5 + & young & 11.5 & 11.5 + + @xmath49 @xmath45 @xmath5 @xmath45 @xmath48 & old & 12.2 & 12.2 + & young & 12.0 & 11.9 + @xmath50sag2 .
+ @xmath51de lucia & blaizot ( 2007 ) .
+ [ tabla_mean_mvir_mr ] figure [ scatter_mass_stmass ] shows that the scatter between virial mass and the stellar mass is similar to that using @xmath5 .
therefore , there is little difference in using luminosity or stellar mass as a proxy for mass ( at least that of the substructure ) .
to adapt the procedure of paper i to galaxies , we need to understand the age - density aspect of the assembly bias in both simulations and observations .
figure [ deltas ] shows a relative stellar age map in our simulation , where the relative age @xmath18 is calculated with respect to virial mass using equation ( 2 ) of paper i. the colours represent different average values of @xmath18 in grids conformed by cells of 2 @xmath27 mpc on a side in a slice 5 @xmath27 mpc thick of our simulation . in general ,
regardless if the relative stellar age is calculated with respect to virial mass o luminosity , values of @xmath52 0.05 correspond to relatively old populations , while @xmath53 -0.05 to young populations .
as can be seen , the latter appear less clustered than the old population even after , in this case , the mass dependence is removed using the definition of @xmath18 .
= 8.9 cm we quantify this behaviour using the @xmath4 index .
the distribution of @xmath4 for our sdss galaxies is shown in the left - hand panel of figure [ distribdn4000 ] for different ranges of absolute magnitude .
we see a clear bimodality for all of the magnitude bins plotted except for the brightest one , probably because most of these objects are part of the ` red sequence ' ( poor gas galaxies with low star formation rate ) . the vertical - dotted line shows the cut that divides the two populations ; active galaxies ( @xmath54 ) and quenched galaxies ( @xmath55 ) .
note that it is very similar to the condition used by tinker , wetzel @xmath13 conroy ( 2011 ) , where quenched objets have @xmath56 .
the right - panel of fig .
[ distribdn4000 ] shows the fraction of quenched sdss galaxies as a function of the large - scale environment .
the latter is defined by counting the number of neighbour galaxies inside a sphere of 10 @xmath27 mpc radius around each galaxy ( see muldrew et al .
2012 for a revision of different environment definitions including this one ) .
we also restrict to neighbours above the magnitude limit for each sample .
we see that the fraction of quenched objects increases to higher densities .
our results are qualitatively similar to those found by tinker , wetzel @xmath13 conroy ( 2011 ) using this index .
= 8.4 cm = 8.4 cm the trend that old objects are located in regions of higher densities than young objects of equal mass or magnitude is also readily seen in the two - point correlation function ( see section [ section_two - point_sdss ] , for example ) , where the ones with the higher clustering at large scales are those located on higher density regions , which could be indicating the locations of higher initial density peaks .
the correlation function , @xmath57 , has the advantage that it obtains an accurate characterization of the spatial distribution of objects from kpc scales out to mpc scales by simply measuring the excess of pairs at a distance @xmath17 with respect to a random distribution .
therefore , @xmath57 is a useful tool to study the large - scale environment .
the overall results will support that older populations are located in higher densities compared to younger ones regardless the luminosity .
high density environments play a key role in properties such as the formation age . later in this work
we will use this density - age connection to obtain a peak - height proxy .
we will show that the crowding around objects helps us trace the assembly bias effect as well as the smooth density field approach introduced in paper i.
in this section , we will test whether sdss galaxies show an assembly type bias by using the two - point correlation function for the centres and tracers in the samples of tables [ tabla_dr4_centres ] and [ tabla_dr4_tracers ] . due to the redshift space distortions
, we have to calculate first the cross - correlation in two coordinates @xmath58 , where @xmath59 is perpendicular and @xmath60 is parallel to the line of sight .
figure [ fede_sdss ] shows the resulting @xmath58 for the old and young galaxies in the magnitude range @xmath61 . as can be seen
there is a subtle difference between both populations due to their slightly offset clustering amplitudes ( c.f .
figure [ xi_dr4 ] ) , but also possibly due to different finger - of - god and infall effects . given that the differences in infall are too subtle
, we will not use the these profiles as estimators of the assembly bias in constrast to paper i. where we reproduce the method used in lacerna @xmath13 padilla ( 2011 ) . ]
= 7.7 cm = 7.7 cm = 8.4 cm = 8.4 cm = 8.4 cm = 8.4 cm i use the jackknife technique to estimate the variance error in the calculation of @xmath58 . in particular the original data is divided in 20 subsamples .
the correlation function is calculated by discarding one of the subsamples at a time , and then it is compared with @xmath58 obtained for the whole sample in order to build the error estimator .
we integrate over the @xmath60 component to obtain the projected correlation function @xmath62 , @xmath63 + where @xmath64 = 0.1 @xmath1 and @xmath65 = 30 @xmath1 .
this can be inverted to obtain the real - space cross - correlation function using the relation described by ( 1992 , see also paz , stasyszyn @xmath13 padilla 2008 ) , @xmath66 + the panels of figure [ xi_dr4 ] show @xmath57 for old and young objects , selected using ranges in @xmath18 , for samples of different luminosities .
error bars are calculated taking into account the error estimate for @xmath58 and using equations ( [ eq_xi_proj ] ) and ( [ eq_xi_inverted ] ) .
we systematically find that the old galaxies show a higher clustering relative to the young population at large scales ( @xmath67 @xmath27 mpc ) for different ranges in @xmath5 .
this difference typically increases toward brighter objects from a 25 per cent to a 45 per cent effect .
therefore , sdss galaxies of equal luminosity show a significant dependence of clustering amplitude with the age since the differences are over a 20 per cent effect . as can be seen from table [ tabla_mean_mvir_mr ] , the mass - to - light ratio @xmath68 is not higher for old galaxies at fixed luminosity in the simulation , so that this dependence of clustering amplitude with age is the assembly bias effect .
this result is in agreement with wang et al .
( 2008 ) , zapata et al .
( 2009 ) and cooper et al .
( 2010 ) who find an assembly - type bias in galaxies .
other works , however , mention that the environmental dependence in galaxies is not important at large - scales . for example ,
tinker , wetzel @xmath13 conroy ( 2011 ) found that the fraction of quenched galaxies increases with the large - scale environment , but this arises just from the dependence of the halo mass on the environment .
they use the @xmath4 parameter to classify a quenched galaxy ( @xmath56 ) .
we find that when using this index to separate active / young and quenched / old objects , the difference in clustering amplitude at large scales due to assembly bias is not clear .
the left - panel of figure [ dn4000ratio ] shows the ratio of the correlation function of old galaxies to the total sample of a given magnitude and the same ratio but using young galaxies , where the determination of age is based on @xmath4 in equation ( [ eq_delta ] ) .
= 8.4 cm = 8.4 cm the difference in the clustering amplitude between the old and young population is very noisy using this estimator for the age , in constrast to the results obtained using the luminosity - weighted stellar age shown in the right panel ( see also lower boxes of fig .
[ xi_dr4 ] ) .
therefore , the @xmath4 index is less adequate to show clearly the environmental dependence at large scales , even when we have 12 per cent more galaxies with this index than with the stellar age ( c.f .
section [ section_sdss_gal ] ) . probably , this is due to the fact that the @xmath4 index is related with star formation by recent bursts in contrast to the luminosity - weighted age which covers an important range of star formation histories .
index among other spectral features to constrain the luminosity - weighted age . ]
we use the @xmath69 = 0 model galaxies and then select four samples of centres and tracers in the same ranges of @xmath5 indicated in tables [ tabla_dr4_centres ] and [ tabla_dr4_tracers ] , respectively .
figure [ xi_sim ] shows the results of the different cross - correlation functions .
= 8.4 cm = 8.4 cm = 8.4 cm = 8.4 cm error bars are calculated using the jackknife method , which has been shown as a robust error estimator for cosmological samples . in general
, this method consists in dividing the sample in parts .
the correlation function is calculated by discarding one of the subsamples at a time , @xmath70 , and then it is compared with the correlation function obtained for the whole sample @xmath57 as follows @xmath71 ^ 2 \textrm { , } \label{eq_jackknife}\ ] ] where @xmath12 is the total number of subsamples . in particular , samples selected from the numerical simulation are divided in ten parts , @xmath72 ( for samples selected from sdss data , @xmath73 ) . as can be seen from these panels , old galaxies ( red circles )
are more strongly clustered than the young objects ( blue squares ) , with average differences of up to a factor of two in the clustering amplitude at large scales , @xmath0 1 @xmath27 mpc . in order to compare the results from real and simulated galaxies ,
figure [ ratios ] shows the ratio @xmath57 of the old galaxies to that of the young galaxies from the simulation and sdss - dr4 ( solid and dash lines , respectively ) .
both ratios are similar for distances beyond 3 @xmath27 mpc , especially for brighter magnitudes .
the clustering of old objects is typically thirty per cent higher than that of young galaxies . at lower scales the ratio is higher for the semi - analytic galaxies , while it remains constant for the sdss galaxies .
this difference may be explained , for instance , by the simplified method of circular orbits in the simulation that could affect the profiles on the one - halo term ( lagos , cora @xmath13 padilla 2008 ) .
note that guo et al .
( 2011 ) found that the projected autocorrelation functions for their semi - analytic galaxies are higher than for observations at scales @xmath74 mpc .
they suggest what this is due to the fact that their assumed fluctuation amplitude , @xmath75 , the same value used in this work , is too high .
the latest results of wmap7 @xcite estimate @xmath76 .
furthermore , figure [ ratios ] shows the ratios of two correlation functions , pointing to a stronger overprediction for old galaxies which reside in denser regions , which is consistent with guo et al .
( 2011 ) , who also find that the simulation overpredicts the autocorrelation functions for passive galaxies , while underpredicts it for active galaxies .
in paper i , we presented an approach to trace the assembly bias effect .
this consisted in measuring the total mass inside spheres of different radii around semi - analytic galaxies , which in some cases could be larger than the virial mass of the host dark matter halo . using two free parameters to introduce a dependence of this radius on mass and age , this model defined a new overdensity peak height for which the large - scale clustering of objects of a given mass did not depend on the age .
we then analyzed the changes of this peak height proxy , compared to virial masses .
we typically found that low - mass , old galaxies have larger peak heights than young galaxies of equal virial mass , which implies that the environment is more dense for old objects .
they are preferentially surrounded by massive neighbour haloes , which is in agreement with other works that mention old , low - mass haloes may suffer a truncated growth due to massive neighbour haloes ( wang , mo @xmath13 jing 2007 ; dalal et al .
2008 ; hahn et al .
therefore , the parametrization of the radius for the sphere to measure the environment is a useful tool to understand what is behind the assembly bias .
however , it is difficult to obtain the underlying mass information for real galaxies .
for this reason , the aim of this section is to design a similar procedure , but using the luminosity of galaxies .
we take advantage of the fact that galaxies of equal virial mass have similar luminosities ( c.f .
section [ section_sam ] ) .
the idea behind our work is not to minimize the bias , but it is to use this method to understand the physical features that produce the assembly bias . instead of using spheres as in paper i , we measure the total luminosity ( which correlates with total substructure mass ) inside a cylinder centred on each galaxy along the line - of - sight to take into account the limitations imposed by the redshift space distortions .
the radius of the cylinder for each galaxy in @xmath27 mpc units is parameterized as @xmath77 + where the characteristic magnitude is @xmath78 @xcite .
the free parameters are @xmath79 and @xmath80 .
the length of the cylinder is @xmath81 = @xmath82 500 km / s centred in each galaxy .
500 km / s corresponds to a total length of 5 @xmath1 in the @xmath69-coordinate of the simulation .
] we add the luminosity @xmath83 of these objects to obtain the total luminosity in this volume using the relation @xmath84@xmath85 .
when the radius is negative the new magnitude of the central object remains unchanged , i.e. @xmath86 .
if this radius is greater than 3 mpc , the object will not be considered for further analysis ( for the best - fitting set of parameters , none of the galaxies exceed this limit in radius ) .
notice that this cylinder will include neighbour galaxies and therefore this will be related to the crowding of the environment of the galaxy , a measure of the local density , but with an adaptive volume size depending on the galaxy age and luminosity .
this procedure to trace the assembly bias will use correlation functions of relatively old and young galaxies to fit the parameters of the cylinder radius , but this parametrization will only be said to correlate with the correction of paper i ( missing peak mass ) if the excess of crowding in galaxies correlates with the excess of mass found in haloes reported in our previous paper . in order to demonstrate that this is the case ,
we first need to find the parameters that trace the assembly bias using this measurement of crowding , and then to check whether the relation between crowding and missing peak mass is present .
in contrast to paper i , we do not use the infall velocity profile because the velocity inferred from redshift space distortions is noisy compared to the velocity from numerical simulations .
our aim is to affect the model with the observational problems for which we define the @xmath69-axis of the simulation as the line - of - sight and turn the @xmath69-coordinate into a recession velocity , @xmath87 , where @xmath88 is the velocity component in the @xmath69-coordinate .
when measuring the luminosity in each cylinder in the numerical simulation , we include all the objects with stellar masses higher than @xmath89 @xmath27 m@xmath30 and magnitude @xmath5 brighter than the limit given for each sample in table [ tabla_dr4_tracers ] .
we calculate the cross - correlation functions between samples of a given new magnitude @xmath90 and tracer galaxies that satisfy @xmath91 .
we also select subsamples of old and young objects using the proxy for the age , @xmath18 , as measured in section [ section_age_parameter ] but using @xmath90 . in order to find the best - fitting parameters that do not show an age dependence of the clustering
, the reduced @xmath92 for the correlation function statistics is defined as @xmath93 ^ 2}{\delta^2_{n(r)}}\right)_i \textrm { , } \label{eq_chixi}\ ] ] + and is calculated within the range 0.8 @xmath94 mpc @xmath95 , mostly in the two - halo term ; @xmath96 is the number of neighbours for young galaxies , whereas @xmath97 is the same quantity for old objects .
the number of neighbours is defined as @xmath98 @xmath99 @xmath100 , where @xmath101 is the mean number of tracers ( see paper i for details ) .
the error is @xmath102 , calculated as the relative error of the number of neighbours .
the symbol @xmath103 denotes the number of degrees of freedom .
the value @xmath104 is the average over four magnitude bins selected between the 10 and 70 percentiles of the @xmath90 distribution .
this particular definition of goodness of fit avoids selecting parameters favoured by large uncertainties that can induce spuriously good fits .
= 8.9 cm = 8.9 cm = 6.5 cm = 6.5 cm = 7.9 cm the best - fitting parameters corresponding to the highest likelihood in the full parameter space are @xmath79 = 0.75 and @xmath80 = @xmath370.36 ( @xmath105 0.51 ) .
figure [ ratios_sincorr ] shows the ratios of the correlation functions of old and young objects using these values for different magnitude ranges .
the assembly bias is practically absent in the numerical simulation using this formalism ( compare with solid lines in figure [ ratios ] ) .
the average differences in clustering amplitude are smaller than a 10 per cent for each range in @xmath90 .
note that we do not include objects with fainter magnitudes than the limit given by the central galaxy in the cylinder .
figure [ comp_mr_prima ] shows the true total magnitude if we also include these neighbour galaxies vs. @xmath90 . there is a strong relation between both magnitudes with a small scatter that is more important for fainter objects .
furthermore , if we use all the neighbour galaxies , ignoring the restriction of the flux limited catalogues , the best - fitting parameters are @xmath106 and @xmath7 ( @xmath107 ) ; the parameter @xmath79 changes slightly to a lower value , while the parameter @xmath80 is the same .
figure [ likelihoods ] shows the likelihood functions for the no - flux limit approach ( dashed lines ) and those of the magnitude @xmath90 ( solid lines ) .
notice that the maximum likelihood in this figure occurs for parameter values that do not necessarily coincide with those of the maximum likelihood in the full parameter space .
we find that both approaches for the luminosity in the @xmath17-band obtain similar maximum likelihood values consistent within a 1@xmath59 confidence ( indicated by the intersections of the parameter distributions with the dotted line ) .
this justifies our use of @xmath90 as a good estimator of the total luminosity around galaxies in the range @xmath16 with the best - fitting values @xmath79 = 0.75 and @xmath80 = @xmath370.36 , avoiding the complications introduced by the flux limit nature of the observational catalogues . in the next section
we show a similar approach to redefine the overdensity peak height using luminosities applied to sdss galaxies .
figure [ vol_limited_mprima ] illustrates the volume - limited samples used in this section in order to avoid obervational bias , where @xmath90 is the redefined luminosity .
we have hundreds of volume - limited samples such as figure [ vol_limited_mprima ] given by the different combinations of the parameters @xmath79 and @xmath80 , thus this was done automatically in order to optimize the number of objects in selected subsamples in luminosity .
the procedure to define these subsamples is as follows . for each cut in @xmath90
, we count the number of galaxies in four ranges of redshift indicated in the last column of table [ tabla_dr4_centres ] .
we then select the range that is most populated that defines faint and bright limits in @xmath90 and a maximum redshift , @xmath108 .
all the galaxies located between the magnitude limits of this range with @xmath109 are selected as members of a centre sample , and all the galaxies brighter than the faint - end limit of this range with @xmath109 are selected as a tracer sample in order to perform cross - correlation functions .
for example , the samples shown in figure [ vol_limited_mprima ] correspond to the parameters @xmath110 , @xmath111 . in this case , for the first cut in magnitude , we select the galaxies with @xmath112 and @xmath113 ( red dots ) as the first centre sample .
its respective tracer sample are the galaxies that satisfy @xmath114 and @xmath113 . for the last cut in magnitude , we select the centre sample ( cyan dots ) using the conditions @xmath115 and @xmath113 .
galaxies for the respective tracer sample satisfy @xmath116 and @xmath113 .
we again select subsamples in @xmath18 using the distribution of the luminosity - weighted stellar age as a function of @xmath90 .
the correlation functions are obtained in redshift space as shown in figure [ fede_best ] .
we estimate their real space counterparts as described in section [ section_two - point_sdss ] .
the number of neighbour objects in equation ( [ eq_chixi ] ) using sdss samples is at least an order of magnitude lower than that in the simulation , so that the statistic to perform the reduced @xmath92 is affected by significant noise .
the number of neighbours is typically smaller than dozens of sdss objects , whereas this number is greater than a hundred for simulated objects .
this is due to the way the number of neighbours is calculated in equation ( [ eq_chixi ] ) , which is proportional to the mean number of tracers at a distance @xmath17 , @xmath101 @xmath117 , where the total volume of the simulation is @xmath118 mpc@xmath28 and , for example , the total volume covered by the sdss sample at @xmath119 is @xmath120 mpc@xmath28 .
for this reason , the reduced @xmath92 in this case ( @xmath90 for sdss galaxies ) is defined as @xmath121 ^ 2}{\delta^2}\right)_i \textrm { , } \label{eq_chixi_spinchapter}\ ] ] where @xmath122 and @xmath123 is the result for old and young sdss galaxies , respectively .
the error is @xmath124 , where the first term is the jackknife error for @xmath122 and the second for @xmath123 .
this statistics is estimated within the range 0.7 @xmath94 mpc @xmath95 and averaged over the four magnitude bins selected in @xmath90 . for the sdss galaxies , the best - fitting values are @xmath8 , @xmath7 .
the marginalized probability distributions for @xmath79 and @xmath80 are shown as dot - dashed lines in figure [ likelihoods ] , and as can be seen there is a good agreement with the distributions obtained from our simulation .
it is remarkable that the parameter @xmath79 has a value only slightly greater compared to that obtained for the galaxies in the simulation ( @xmath6 ) , which lies within the uncertainties of this parameter , and the value of @xmath80 is equal in both cases .
the correlation function ratios using these values are shown in figure [ sdss_noassembly ] .
again , the assembly bias is almost not present using the proxy for the overdensity peak height at @xmath0 1 @xmath27 mpc ( average differences in clustering amplitude are smaller than a 10 per cent for each range in magnitude @xmath90 ) , thus confirming that our approach is able to define an initial peak - height proxy using the crowding of the region where galaxies lie , in a similar manner to that implemented in paper i , and that sdss galaxies show a similar behaviour to @xmath26cdm semi - analytic galaxies since their best - fitting sets of parameters are comparable .
= 7.7 cm = 7.7 cm in this context , as mentioned above , figure [ likelihoods ] shows the marginalized likelihood distributions for the free parameters @xmath79 and @xmath80 of equation ( [ eq_r_prm ] ) for three cases : the new magnitude @xmath90 for simulated galaxies , the true total magnitud in the @xmath17-band using also simulated galaxies and @xmath90 for sdss galaxies . by estimating a weighted average for the values in the range of @xmath9 for the three distributions , it is obtained @xmath125 @xmath126 + the best estimates for the @xmath79 parameter
are always positive , whereas those for the @xmath80 parameter are negative when defining an overdensity peak height proxy using luminosities .
the next section will discuss the properties of this proxy of peaks when using galaxy luminosities .
the redefinition of the overdensity peak height using galaxy luminosities is able to trace the assembly bias ( section [ sec_prm ] ) .
table [ tabla_radius_lum_lcp ] shows the median radius of equation ( [ eq_r_prm ] ) in physical units of @xmath27 kpc for the four different bins in @xmath5 using the simulated galaxies .
the results are shown for two sets of best - fitting values , the @xmath90 case ( @xmath127 , @xmath128 ) and the true total magnitude in the @xmath17-band ( @xmath110 , @xmath128 ) .
the radius of the cylinder for all the objects decreases to higher luminosities with very similar sizes for both sets of parameters .
this is in agreement to what was discussed in paper i , where the radius decreases to more massive objects , due to the correlation between luminosity and virial mass .
bear in mind that the radius can be negative , @xmath129 , since both the first and second term of equation ( [ eq_r_prm ] ) can be negative , which implies @xmath86 .
the results are also split according to the classification of old and young populations after performing the redefinition of luminosity for these simulated galaxies . as well as
paper i , old objects have higher radii than young objects ( of equal luminosity in this case ) , thus suggesting stronger enviromental dependences for old populations .
the median sizes for both sets of parameters are similar for young galaxies , with differences smaller than 37 @xmath27 kpc , while these differences for old galaxies range from 8 @xmath27 kpc out to 200 @xmath27 kpc toward brighter magnitudes .
= 8.9 cm c c c c c c c + best - fitting parameters & ages & @xmath130kpc@xmath131 & @xmath130kpc@xmath131 & @xmath130kpc@xmath131 & @xmath130kpc@xmath131 + & & @xmath5 = @xmath132 & @xmath5 = @xmath133 & @xmath5 = @xmath134 & @xmath5 = @xmath135 + @xmath79 = 0.75 , @xmath111 & all & 716.4 & 380.6 & 29.1 & @xmath45 0 + & old & 910.7 & 723.1 & 523.4 & 206.6 + & young & 582.2 & @xmath45 0 & @xmath45 0 & @xmath45 0 + + @xmath79 = 0.50 , @xmath111 & all & 719.9 & 385.0 & 28.4 & @xmath45 0 + & old & 902.8 & 596.7 & 340.1 & 6.5 + & young & 593.8 & 36.6 & @xmath45 0 & @xmath45 0 + [ tabla_radius_lum_lcp ] c c c
c c c c + best - fitting parameters & ages & @xmath130kpc@xmath131 & @xmath130kpc@xmath131 & @xmath130kpc@xmath131 & @xmath130kpc@xmath131 + & & @xmath5 = @xmath132 & @xmath5 = @xmath133 & @xmath5 = @xmath134 & @xmath5 = @xmath135 + @xmath79 = 1.0 , @xmath111 & all & 694.6 & 358.3 & 28.2 & @xmath45 0 + & old & 851.8 & 726.1 & 535.7 & 294.5 + & young & 606.1 & @xmath45 0 & @xmath45 0 & @xmath45 0 + [ tabla_radius_lum_sdss ] table [ tabla_radius_lum_sdss ] shows the median radius of equation ( [ eq_r_prm ] ) using the sdss galaxies . in general , there is a good agreement of the radius sizes with respect to the @xmath90 case ( see table [ tabla_radius_lum_lcp ] ) , with differences among them smaller than 22 @xmath27 kpc for all the galaxies , 88 @xmath27 kpc for the old galaxies , and 24 @xmath27 kpc for the young galaxies .
this means that the best - fitting set of parameters that make the assembly bias practically absent for both simulated and real galaxies are truly similar when using luminosity as a proxy of peak height .
= 8.9 cm the method followed to redefine the luminosity of a galaxy could be compared to adding up the mass from the smooth density field beyond the virial radius .
we compare the resulting @xmath90 luminosity function with that of the total luminosity in haloes in the simulation to help interpret the changes introduced by our approach ; this is equivalent to our procedure in paper i where we compared the mass functions for the original virial mass , and that resulting from our redefinition of peak height , although in this case the luminosity function does not have a theoretical prediction .
figure [ lf_m9.1 ] shows differences in the luminosity function that results from the redefinition of luminosity @xmath90 for central galaxies ( solid circles ) and that from adding the luminosity in the @xmath17-band of all the galaxies with @xmath91 in the same friends - of - friends ( fof ) halo ( open triangles ) as an approximation of the total luminosity within the peak .
the latter is defined for fof haloes where the magnitude of their central semi - analytic galaxy is @xmath136 , otherwise the magnitude of fof haloes is simply that of their central galaxy .
as our formalism adds luminosities of galaxies embedded in haloes , one could have expected the number of these objects with magnitudes @xmath90 to be similar to that of all the galaxies in fof haloes , but the difference between the luminosity functions measured with @xmath5 and @xmath90 is considerable . in this section we will study the origin of this discrepancy .
in paper i we found that the redefinition based on the mass density field does not change the halo mass function at @xmath137 @xmath27 m@xmath30 ( see their fig .
8) when compared to that of sheth , mo , & tormen ( 2001 , smt ) or that obtained from the virial mass , which is in agreement with the result found by gao , springel & white ( 2005 ) where the assembly bias with respect to formation time is not important for the high mass regime .
the reason for the change found in the low - mass regime is that the peak for an old , low - mass object adds more haloes and mass than for a young object of equal mass . for the following test , instead of considering the mass from dark matter ( dm ) particles inside a sphere , which corresponds to a continuous field , we use the mass of dark matter haloes in spheres , which is a discrete formalism similar to using the luminosity of individual galaxies .
bear in mind that both estimators measure the neighbours enclosed within the sphere calculating the crowding of the region . in the top and middle panels of figure [ xi_vel_mf_test2 ] we show the correlation function and the velocity infall profile for old and young objects of a given mass for the best - fitting parameters @xmath138 and @xmath139 in equation ( 14 ) of paper i , resulting from using this discrete approach with neighbour dm haloes . compared to the best - fit parameters of paper i ( @xmath140 and @xmath141 ) , these are significantly different .
the difference between the approach for haloes and that used for a smooth density field is evident when looking at the mass function , shown as open triangles in the bottom panel , which is completely different than that using the virial mass ( green squares ) as well as the predicted mass functions from the extended press - schechter model ( eps ) and from the smt model shown as long - dashed and dot - dashed lines , respectively .
we also compare it with a set of parameters with good - fitting results to trace the assembly bias using the continuum formalism ( @xmath140 and @xmath141 , open circles ) .
the behaviour of the latter mass function is in agreement with the virial mass , and the eps and smt models .
= 7.2 cm the discrete formalism actually measures the crowding around objects which constitutes a different proxy of the peak height than the total mass .
figure [ comp_mass ] compares the result obtained with the best - fitting set of values @xmath142 for the luminosity of individual simulated galaxies and that obtained in paper i , which considers the mass of dm particles .
there is a clear correlation between both methods , showing that the same phenomenon is being described .
we calculate a correlation coefficient using the pearson s product - moment coefficient , @xmath143 + where @xmath144 is the difference between the original magnitude @xmath5 and the new one @xmath90 ( y - axis of fig . [ comp_mass ] ) , @xmath145 is its average using the @xmath12 objects , @xmath146 = log(@xmath147 ) is the new mass in units of the original virial mass ( x - axis of fig .
[ comp_mass ] ) , and @xmath148 is its average .
we obtain @xmath149 , which means that there is a strong correlation between both formalisms .
we conclude that although our method to measure the total luminosity will not reproduce the luminosity function , the reason for this is only the use of discrete neighbours .
if we were able to measure the distribution of mass around galaxies and turn it into luminosity , we would obtain a more physical luminosity function , as we did in the case of the mass .
this result implies that the phenomena behind the assembly bias , namely , the truncation of mass infall in haloes of low mass embedded in high density regions , can be detected with the new proxy for peak height that can be applied to real galaxy surveys .
we have shown that the assembly bias effect , namely , the difference in the clustering amplitude at large scales of populations of equal mass or luminosity but different age , exists for both semi - analytic galaxies and real galaxies from a sdss catalogue .
this is a relevant issue that could affect the ability of the next generation of galaxy surveys to infer accurate cosmological parameters . using the two - point correlation function ,
the clustering strength of old sdss galaxies is from a 25 per cent to a 45 per cent higher than that of young sdss galaxies of same magnitude at scales @xmath67 @xmath27 mpc .
the estimator of age is based on the mass- and luminosity- weighted stellar ages for the synthetic and sdss galaxies , respectively . in this context
, we show that the @xmath4 index is a less adequate parameter to separate active / young and quenched / old galaxies for assembly bias studies using correlation functions at large scales .
lacerna @xmath13 padilla ( 2011 , paper i ) presented an overdensity peak height proxy with the aim to understand the assembly bias effect .
this new definition was proposed as a better alternative than the virial mass . in this work
we adapt this model including observational issues using galaxy luminosities , taking advantage of their correlation with substructure mass . as the sdss
is limited in flux , and as the tracers of mass are the galaxies themselves , the method is modified to use the crowding of the environment around galaxies in selected volume - limited samples , instead of searching for a correction to the virial mass as was done in paper i. we measure the total light in cylinders centred in each galaxy , thus obtaining an estimate of the crowding around galaxies , that traces the asembly bias .
the radius of this cylinder is parameterized as a function of stellar age and luminosity .
the best - fitting sets of parameters in the simulation and observations are very similar ( @xmath6 , @xmath7 and @xmath8 , @xmath7 , respectively ) . in both cases
the asembly bias is practically absent using this approach .
the average differences in clustering amplitude between old and young populations for both semi - analytic and observational galaxies are smaller than a 10 percent for each range in magnitude @xmath90 . by estimating a weighted average for the values in the range of @xmath9 for semi - analytic and sdss galaxies
, it is obtained that the set of best - fitting parameters for the radius of the cylinder is @xmath10 , @xmath11 .
this latter model , which constitutes a discrete formalism when counting galaxies within a cylinder or haloes within spheres , does not reproduce the original luminosity or mass functions since it actually measures the crowding around galaxies .
however , it is correlated with the smooth density field approach of paper i. therefore , both formalisms are associated with the fact that the mass of peaks that did not collapse on to haloes will not be an appropiate proxy for the peak height .
it is likely that low - mass haloes in high density environments show their growth truncated by the gravitational effects of their ( massive ) neighbours .
our results indicate that this also affects the growth of the stellar mass in galaxies in addition to the growth of dark matter haloes .
this could induce a misclassification of galaxies in theoretical models to populate haloes , such as hod , because using virial mass instead of proper peak height .
these results will be of high importance for the next generation of galaxy surveys , such as lsst .
the huge amount of information available will allow us to measure the clustering amplitude of galaxies with significant statistics at mpc scales .
therefore , as we show in this paper , it will be absolutely necessary to model the effect of the environment on the two - halo regime in order to understand its role in the formation and evolution of haloes and galaxies .
theoretical models will have to include this phenomenon in order to reproduce the observational data with high accuracy in the new era of precision cosmology .
we would like to thank harold francke and alvaro orsi for comments and discussions .
we acknowledge support from fondap centro de astrofsica " @xmath150 , basal - cata , fondecyt grant no . 1110328 , conicyt , and mecesup .
support for il is provided by the ministry for the economy , development , and tourism s programa iniciativa cientfica milenio through grant p07 - 021-f , awarded to the milky way millennium nucleus .
fs thanks supports by the dfg research unit for 1254 .
the calculations for this work were performed using the geryon cluster at the centre for astro - engineering at uc .
funding for the sdss and sdss - ii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , the u.s .
department of energy , the national aeronautics and space administration , the japanese monbukagakusho , the max planck society , and the higher education funding council for england .
the sdss web site is http://www.sdss.org/. the sdss is managed by the astrophysical research consortium for the participating institutions .
the participating institutions are the american museum of natural history , astrophysical institute potsdam , university of basel , university of cambridge , case western reserve university , university of chicago , drexel university , fermilab , the institute for advanced study , the japan participation group , johns hopkins university , the joint institute for nuclear astrophysics , the kavli institute for particle astrophysics and cosmology , the korean scientist group , the chinese academy of sciences ( lamost ) , los alamos national laboratory , the max - planck - institute for astronomy ( mpia ) , the max - planck - institute for astrophysics ( mpa ) , new mexico state university , ohio state university , university of pittsburgh , university of portsmouth , princeton university , the united states naval observatory , and the university of washington .
the millennium simulation databases used in this paper and the web application providing online access to them were constructed as part of the activities of the german astrophysical virtual observatory . | we analyse catalogues of galaxies from the sloan digital sky survey ( sdss ) and find a weak but significant assembly - type bias , where old galaxies have a higher clustering amplitude at scales @xmath0 1 @xmath1 than young galaxies of equal magnitude . when using the definition of age based on the luminosity - weighted stellar age , the difference increases from @xmath2 to @xmath3 sdss galaxies , from a 25 per cent to a 45 per cent effect .
in contrast , when using the definition based on the @xmath4 index , the clustering amplitude at large scales shows a weaker signal - to - noise for the assembly bias detection . in semi - analytic galaxies
the difference in clustering , using the stellar age and luminosity , is similar at distances beyond 3 @xmath1 , but they overpredict the assembly bias found in sdss galaxies at smaller scales .
we then adapt the model presented by lacerna & padilla ( paper i ) to redefine the overdensity peak height , that traces the assembly bias such that galaxies in equal density peaks show the same clustering regardless of their age , but this time using observational features .
the proxy of peak height for observational data consists in the luminosity inside cylinders of different radii centred in each galaxy .
this radius is parameterized as a funcion of stellar age and magnitude @xmath5 .
the best - fitting set of parameters that make the assembly bias practically absent for both simulated and real galaxies are similar ( @xmath6 , @xmath7 and @xmath8 , @xmath7 , respectively ) . by estimating a weighted average for the values in the range of @xmath9 for semi - analytic and sdss galaxies
, it is obtained that the set of best - fitting parameters for the radius of the cylinder is @xmath10 , @xmath11 .
the idea behind the parametrization is not to minimize the bias , but it is to use this method to understand the physical features that produce the assembly bias effect . even though in this case we consider neighbour galaxies in cylinders in contrast to using a smooth density field , e.g. the dark matter particles distribution , our simulations show that the new method is correlated to the definition of paper i as both measure the crowding around objects .
[ firstpage ] cosmology : large - scale structure of universe - cosmology : dark matter - galaxies : statistics - galaxies : formation |
young stars are characterized by a large degree of activity .
winds , jets and outflows are manifestations of the interaction of the stars with the circumstellar environment in which they are embedded .
similarly , both t tauri and herbig ae / be stars show photometric and spectroscopic variability on time scales of minutes to years , indicating that photospheric activity begins in the earliest phases of stellar evolution , prior to the arrival on the main sequence ( ms ) . in many cases ,
strong and rapid line variability has been detected with periods close to the star rotation period , interpreted as due to the rotational modulation of spots on the stellar surface or of fast and slow streams in the stellar wind ( e.g. gahm et al .
1995 ; catala et al .
1989 ) . on the other hand , the fact that young stars during their evolution to the ms move across the instability region of post - ms stars raises the possibility that at least part of the activity could also be due to stellar pulsations ( see baade & stahl 1989 ; kurtz & marang 1995 )
however , the instability properties of young stars have not been studied so far and it is not at all clear whether a pre main - sequence star can indeed pulsate .
as recently discussed by gautschy & saio ( 1996 ) , stars over essentially the whole mass spectrum can become pulsationally unstable during various stages of their evolution .
in particular , @xmath3 scuti stars are intermediate - mass variables of spectral type a to f with pulsation periods less than @xmath4 and light amplitudes ranging from thousandths of magnitude to some tenths ( breger 1979 ; breger & pamyatnykh 1998 ) . in spite of the traditional association of @xmath3 scuti pulsation with stars in their core hydrogen - burning phase or evolving towards the base of the giant branch and burning hydrogen in a shell , some @xmath3 scuti have been identified to be pms objects ( breger 1972 , kurtz & marang 1995 ) .
kurtz & marang ( 1995 ) have observed the herbig ae star hr 5999 searching for pulsational photometric light variations characteristic of @xmath3 scuti stars .
they have found a peak - to - peak pulsation amplitude of about @xmath5 in johnson v - band and a period of 4.99 hr , superimposed on a nonperiodic photometric variation of @xmath6 probably due to variations of dust obscuration in the circumstellar envelope and disk .
there are other stars in which evidence for pulsational variability has been advocated to explain the observed rapid line profile variability ( e.g. hd 163296 and hd 250550 , baade & stahl 1989 ) , but the case is not as convincing as that provided by hr 5999 . in any case , the discovery of pulsation in pms stars is extremely important since it provides a unique means for constraining the internal structure of young stars and for testing evolutionary models . in this letter
, we explore the pulsational behavior of stellar models distributed along the evolutionary tracks of pms stars computed by palla & stahler ( 1993 ) .
we identify the boundaries of the instability strip by using nonlinear models .
we can then use these models to constrain the fundamental stellar parameters of hr 5999 and to select a sample of known pms stars that are likely candidates for showing pulsation variability in their photometric light curves .
the calculation of the pulsational instability requires the specification of the stellar parameters , luminosity and effective temperature , for a given mass . we have used the pms models for low- and intermediate - mass stars computed starting at the birthline determined by the protostellar accretion phase ( palla & stahler 1990 , 1993 ) .
the development of protostar theory has altered the concept of pms evolution particularly for stars more massive than solar . in comparison with classical evolutionary tracks that used arbitrary initial conditions ,
those of palla & stahler ( 1993 ) occupy a much reduced portion of the hr diagram as a consequence of the modest stellar radii attained by each star during the accretion period .
in addition , stars between 2 and 4 m@xmath0 are not initially convective from surface cooling , but begin their evolution in a thermally unrelaxed state corresponding to luminosities and effective temperatures lower than those of classical evolutionary tracks .
this sequence of events may have direct effects on the study of the instability properties of pms stars and should be taken into account in the selection of the equilibrium models . as an illustration of the differences in the internal structure of stars at various stages of evolution , fig .
1 shows the variation with radius of the interior density ( normalized to the central value ) for a 3 m@xmath0 star . on the birthline , at the beginning of the pms phase , the density profile is quite flat and the star is thermally unrelaxed . at the end of the relaxation phase , in the middle of the pms track ,
the star has already started its contraction to the zams and is more centrally condensed .
the middle curve of fig .
1 is for a model with luminosity l@xmath7=57 l@xmath0 and effective temperature t@xmath8=6900 k at an age of @xmath9 yr .
as we shall see below , such a model falls within the boundaries of the instability strip . finally , the lower curve is for a star that has completed the main - sequence phase and is evolving toward the red giant branch .
the effective temperature is approximately the same as for the pms star , but the luminosity is now l@xmath7=135 l@xmath0 due to the larger radius ( 8.3 r@xmath0 instead of 5.3 r@xmath0 ) .
these conditions correspond to those of a typical , evolved @xmath3 scuti star .
the star is much more centrally condensed because of the complete exhaustion of hydrogen . however , beyond the inert core , the density profile is quite similar to that of the pms star .
thus , a mature @xmath3 scuti star differs from a young one mostly in the inner regions and for its higher luminosity .
these structural changes clearly affect the estimate of the pulsation periods , as we shall see in the following sections .
the theoretical framework for the investigation of the linear and nonlinear pulsation characteristics has been described in several papers ( e.g. bono et al . 1997 and references therein ) and is briefly summarized here .
we have computed several sequences of linear nonadiabatic models at fixed mass covering a wide range of luminosities and effective temperatures .
these models give information about periods and modal stability and provide the static envelope structures for the nonlinear computations .
we adopt an optical depth of the outermost zone of about 0.001 and fix the inner boundary condition at a fraction of the stellar radius varying between 10 and 20% .
the zone closest to the hydrogen ionization region ( hir ) was constrained to t@xmath10 k and 20 mass zones were inserted between the hir and the surface in order to ensure a good spatial resolution of the outermost regions throughout the pulsation cycle . due to the key role played by spatial resolution for firmly estimating
both linear and nonlinear modal stability of higher modes , the envelopes were discretized by adopting a detailed zoning in mass .
the mass ratio between consecutive zones ( @xmath11 ) has been assumed equal to @xmath11=1.1 for temperatures lower than @xmath12 k and to @xmath11=1.2 for higher temperatures . in this way
a typical envelope model , used for both linear and nonlinear computations , is characterized by roughly 20 - 30% of the total stellar mass and by 150 - 250 zones . to model the physical mechanisms which govern the pulsation phenomenon , the @xmath13 and @xmath14 mechanisms in the hydrogen and helium ionization regions ( see cox 1980 )
, we have adopted the opal radiative opacities provided by rogers & iglesias ( 1992 ) for temperatures higher than @xmath15 k and the molecular opacities of alexander & ferguson ( 1994 ) for lower temperatures . a solar chemical composition ( y=0.28 , z=0.02 )
is assumed in all the computations .
for each selected pulsation mode , the fundamental observables that can be derived from the linear nonadiabatic computations are periods and growth rates . in this work
, we limited the study to the first three radial modes of pulsation . a detailed analysis of the sign of growth rates provides an estimate of the location of the blue boundaries of the instability strips for each mode , representing the maximum effective temperatures allowing pulsation at a given luminosity level .
no information is derived about the location of red boundaries because our linear models are completely radiative while the quenching of pulsation at lower effective temperatures is due to the interaction between convective and dynamical motions ( e.g. baker & kippenhan 1965 ) .
although our goal is the definition of the theoretical instability strip , we have specialized the linear analysis to the case of hr 5999 . in particular , we begin by reproducing its observed pulsation period of 4.99 hr for a combination of mass , luminosity and effective temperature consistent with both observations and evolutionary tracks . to this purpose ,
we have computed a grid of linear nonadiabatic models of stars with masses equal to 1.5 , 2 , 2.5 , 3 , 3.5 , and 4 m@xmath16 , luminosities in the range 10 to 140 l@xmath16 , and effective temperatures 4500@xmath17t@xmath189500 k. the linear pulsation code assumes that the stellar models are in complete thermal and hydrostatic equilibrium .
figure 2 shows the location in the hr diagram of the lines of constant period ( @xmath19 hr ) for the three pulsation modes and for different stellar masses . at a given mass
, each line interesects the corresponding evolutionary track in one point that determines the values of l@xmath7 and t@xmath20 to be used in the nonlinear models .
also shown in fig .
2 is the position of hr 5999 with the observational uncertainties .
the distance to the star based on hipparcos parallaxes is @xmath21 pc which gives a luminosity of 78@xmath22 l@xmath0 ( van den ancker et al .
the spectral classification of hr 5999 is a7iii / ive and estimates of the effective temperature for this spectral type vary between @xmath237600 k and @xmath237000 k ( e.g. van den ancker et al . 1997 ; solano & fernley 1997 ) .
as one can see from fig . 2 , only few equilibrium models with the correct period fall within the hr 5999 box .
in particular , the model with mass m@xmath7=3 m@xmath0 pulsates in the fundamental mode , that with m@xmath7=3.5 m@xmath0 in the first overtone , and that with m@xmath7=4 m@xmath0 in the second overtone .
the selection of these models restricts the choice of acceptable stellar masses for the detailed nonlinear analysis .
given the static envelope structures produced by the linear computations , we have performed the nonlinear stability analysis .
the equations governing both the dynamical and convective structures were integrated in time until the initial perturbations and the nonlinear fluctuations , which result from the superposition of higher order modes , settled down ( for more details see bono , castellani & stellingwerf 1995 ) . because of
the very low growth rates ( between @xmath24 and @xmath25 ) , an extremely large period number ( exceeding 10,000 ) must be calculated for each model before reaching the stable limit cycle of pulsation ( maximum amplitude ) .
the initial velocity profile was obtained by perturbing the linear radial eigenfunctions with starting surface velocities varying in the range 0.1 to 10 km s@xmath26 .
as found in investigations of post main - sequence @xmath3 scuti star models ( milligan & carson 1992 ; bono et al . 1997 ) , the modal stability analysis reveals the presence of stable nonlinear limit cycles in the fundamental , first and second overtone as the effective temperature increases . figure 3 shows the location of the theoretical instability strip boundaries in the hr diagram , together with the last static and the first pulsating models found for different masses . the stellar parameters along the strip are listed in table 1 . in agreement with the distribution originally suggested by breger & bregman ( 1975 ) and whith previous theoretical predictions ( bono et al . 1997 ) , we find that second and first overtone pulsators are located at effective temperatures higher than the fundamental ones .
all the models on the red edge are pulsating in the fundamental mode , while those on the blue edge in either of the overtone modes . the _ red _ edge varies between @xmath236500 k and @xmath237100 k , whereas the _ blue _ edge between @xmath237100 @xmath27 and @xmath237500 k. the width of the instability strip is nearly constant and equal to about 650 k. the time spent by each star within the strip is also a constant fraction of the total pms contraction time , typically between 5 and 10% , although in absolute terms it decreases from @xmath2310@xmath28 yr for m@xmath7=1.5 m@xmath0 to 8@xmath2910@xmath30 yr for m@xmath7=4 m@xmath0 . as shown in fig . 3
, the pms instability strip terminates abruptly at the blue edge of the m@xmath7=4 m@xmath0 evolutionary track .
this results is the consequence of our assumption that pms stars begin their contraction phase at the birthline computed with an accretion rate of @xmath31 m@xmath0 yr@xmath26 .
it has been shown that such birthline reproduces quite well the the distribution of known pms stars in the hr diagram ( palla & stahler 1990 ) .
obviously , the theoretical instability strip extends to luminosities higher than the upper limit found here for pms stars , but it would then be populated only by more evolved objects such as the classical @xmath3 scuti stars .
the models presented in this paper represent the first attempt to derive the boundaries of the instability strip for young stars .
a complete analysis of the predicted light curves and amplitudes and of the sensitivity to the input physics will be given elsewhere ( marconi et al .
1998 , in preparation ) . however
, we can use these initial results to set constraints on the pulsational properties of known pms objects .
the location of hr 5999 with respect to the instability strip is shown in panel ( b ) of fig . 3 .
we can see that in order to fall within the strip , the effective temperature of hr 5999 should be lower than @xmath237250 k if its mass is 3 m@xmath0 or equal to @xmath237100 k for m@xmath7=4 m@xmath0
. however , if we accept the pulsation period of 4.99 hr found by kurtz & marang ( 1995 ) , then the only acceptable models would be ( see fig .
2 ) a 3 m@xmath0 star pulsating in the fundamental mode , a 3.5 m@xmath0 in the first overtone , or a 4 m@xmath0 star in the second overtone . on the other hand ,
these three models are located at the blue edge of the strip and the nonlinear stability analysis predicts that they should all pulsate in the second overtone mode .
but in this higher mode only the 4 m@xmath0 model oscillates with the correct period of hr 5999 , ruling out the 3 m@xmath0 and 3.5 m@xmath0 stars as acceptable models .
thus , we conclude that the pulsation properties of hr 5999 suggest a mass of 4 m@xmath0 with second overtone pulsation . despite the relatively short time spent by pms stars inside the strip , fig .
3 shows that there is a number of known objects with the right combination of luminosity and temperature to qualify as candidate pulsators .
these include five stars ( v376 cas , cq tau , lkh@xmath32 208 , hd 142666 and kk oph ) of the ux orionis class , characterized by large photometric and polarimetric variations attributed to variable extinction by circumstellar dust ( grinin & tambovtseva 1995 ; natta et al . 1997 ) . as in the case of hr 5999
, these stars should show periodic variations of smaller amplitudes in their photometric behavior . also , there are eight herbig ae / be stars of spectral type between a7 and f7 ( hd 35929 , v351 ori , hd 142527 , lkh@xmath32 233 , ty cra , rcr a , tcr a and hd 104237 ) which are located inside or near the instability strip within the uncertainties of the determination of l@xmath7 and t@xmath8 ( berrilli et al .
1992 ; van den ancker et al .
1998 ) . searches for @xmath3 scuti type variability in these stars are now under way .
it is a pleasure to thank g. bono and l. terranegra for useful discussions on pulsation models and spectral properties of young stars .
this project was partly supported by cnr grant 97.00018.ct02 and by asi grant ars 96 - 66 to the osservatorio di arcetri .
lccccccccc 1.5 & 7 & 7090 & f & 0.0611 & & & & & 2.0 & 18 & 6870 & f & 0.1195 & & 21 & 7500 & ii & 0.0690 2.5 & 32 & 6640 & f & 0.2155 & & 37 & 7350 & ii & 0.0925 3.0 & 52 & 6500 & f & 0.2601 & & 62 & 7260 & ii & 0.1299 3.5 & 77 & 6550 & f & 0.3195 & & 95 & 7200 & ii & 0.1721 4.0 & & & & & & 133 & 7100 & i & 0.2736 | we investigate the pulsational properties of pre main - sequence ( pms ) stars by means of linear and nonlinear calculations .
the equilibrium models were taken from models evolved from the protostellar birthline to the zams for masses in the range 1 to 4 m@xmath0 .
the nonlinear analysis allows us to define the instability strip of pms stars in the hr diagram .
these models are used to constrain the internal structure of young stars and to test evolutionary models .
we compare our results with observations of the best case of a pulsating young star , hr 5999 , and we also identify possible candidates for pulsational variability among known herbig ae / be stars which are located within or close to the instability strip boundaries .
= cmbx10 scaled = cmbx10 scaled1 # 1;#2;#3;#4 # 1 , # 2 , # 3 , # 4 .
# 1;#2 * # 1 * : _ # 2 _ # 1@xmath1 # 1@xmath2 # 1equation ( [ # 1 ] ) # 1equation ( [ # 1 ] ) # 1 ( [ # 1 ] ) # 110^#1 # 1n _ # 1 # 1k_#1 |
experiment e864 at brookhaven has a quite comprehensive set of measurements which addresses the topic of coalescence of multibaryon states in heavy - ion collisions at ags energies .
these include measurements of stable light nuclei from mass number @xmath4=1 ( including the first measurement of neutrons in an ags au+pb or similar system ) up to @xmath4=7 .
in addition , we have measured production of the strongly decaying states @xmath1 and @xmath2 and the hypernucleus @xmath3 . with this set of measurements we can systematically examine the dependences of production of coalesced states on kinematic variables and collision centrality as well as on the mass number , spin , isospin , and strangeness content of the state .
only for the lighter states ( @xmath5 ) do we currently have sufficient statistics to study the kinematic and centrality dependences of production . in figure [
fig : temp_vs_mass ] we display inverse slope parameters for protons , deuterons , and @xmath6 nuclei as a function of mass number .
these slope parameters are determined from boltzmann fits in transverse mass within the rapidity range from 2.2 to 2.4 and are shown both for 10% most central events and more peripheral events .
polleri et .
@xcite have demonstrated that such trends in the temperature parameters of light nuclei are sensitive to both the spatial profile of coalescing nucleons and to the flow profile of the source .
our reach in mass up to @xmath4=3 is then very useful in untangling these two profiles .
if we focus only on a narrow slice in transverse momentum at @xmath7=100 - 200 mev / c , we can examine the yields of these light nuclei as function of rapidity .
we find that the invariant yields are concave in shape ( i.e. are lowest at midrapidity and increase toward beam rapidity ) and that the relative concavity increases as we move to nuclei with higher masses .
this can be understood as an effect of longitudinal flow or incomplete stopping .
these and other trends in these lighter nuclei are explored much more fully in @xcite .
we now focus only on 10% most central events and on a kinematic region around @xmath8=1.9 , @xmath7=200 mev / c and explore the dependences of production on properties of the nuclei . to start simply , we plot in figure [ fig : adep ] the invariant yields in this narrow kinematic region as a function of mass number @xmath4 for @xmath0 .
we see that over almost ten orders of magnitude , the yields are well described by a simple exponential dependence ; implying that to add a nucleon to a cluster , one pays a penalty factor of approximately 48 .
there are small deviations from this trend , and we next move to trying to explain those . two dependences that are apparent are the spin and isospin dependences of the yields . examining the ratios of invariant yields for @xmath9 , @xmath10 and @xmath11 near this same kinematic range , we find that @xmath9 and @xmath10 are consistent with a value of approximately 1.2 while the @xmath11 ratio is nearer to 0.3 .
the neutron , proton , triton , and @xmath6 are all spin @xmath12=1/2 , while @xmath13 is @xmath12=0 and @xmath14 is @xmath12=1 .
we take this as evidence that the yields of these nuclei include a dependence of @xmath15 ( with @xmath16 ) where the first term accounts for the neutron to proton ratio at freeze - out and the second term is for the spin degeneracy of the state as is generally included in coalescence models .
we next take these measured invariant yields from figure [ fig : adep ] and divide them by @xmath17 to remove these dependences on mass number , spin , and isospin of the states .
we then look for any other trends in these corrected yields .
in fact , we do see such a trend when we plot these ratios as a function of binding energy per nucleon as shown in figure [ fig : bind ] .
this binding energy per nucleon is defined as the sum of all the masses of the constituent nucleons of a state less the nuclear mass all divided by the number of nucleons .
we note the low temperature parameter of 6 mev that we extract from this fit ( strongly decaying states were not included in determining the fit ) .
it is possible then that this represents some final freeze - out of the nucleons at a much lower temperature than kinetic freeze - out , or it is possible that this is evidence of some other trend that coincidently shows up as a correlation with this binding energy .
we are also exploring other possible correlations such as with nuclear size .
to complement these measurements of stable light nuclei , we have made measurements of strongly decaying nuclei and the hypernucleus @xmath3 .
because these analyses both involve states which decay at most centimeters from our target ( and therefore have decay products which are indistinguishable in our apparatus from particles originating from the target ) , they proceed along similar lines .
the measurements are made in each case by identifying both of the decay products in our apparatus .
when a pair of decay products is found in an event , an invariant mass for the pair is calculated and placed in a histogram of pair invariant masses . to mimic and subtract away the background , pairs consisting of one decay product from one event and the second from a separate event
are put though this same process , producing a second invariant mass histogram .
the properly normalized difference of these two histograms should then be the signal under study as shown in figure [ fig : h3l ] ( see @xcite and @xcite for details ) . with this method we measure @xmath2 ( @xmath1 ) through its decay to @xmath18 . on a gross scale ,
the invariant yields of these states sit closely along the trend shown in figure [ fig : adep ] .
but on closer inspection in figure [ fig : bind ] , the yields for these states sit lower by a factor of 2 to 4 than the trend set by the stable light nuclei . perhaps still more interesting
is the absence of any signal for the @xmath19 16.7 mev excited state from its decay to @xmath20 . from the absence of a signal
, we set the 90% confidence level upper limit on production of this state as shown in figures [ fig : adep ] and [ fig : bind ] .
note that if we assume that the ground state production sits at its most probable value from our measurement , then this upper limit implies an upper limit for an excited state temperature of 15 mev in a simple thermal picture ; this is consistent with measurements @xcite made at the lower energies of gsi and possibly indicative @xcite of the same final - state interactions that may give rise to the the trend seen in figure [ fig : bind ] . after the analysis of nearly 2/3 of the relevant data
, we observe a signal for @xmath3 at approximately the @xmath21 level .
when our analysis is complete , we expect if this is indeed a signal that with improved analysis methods and the use of the full statistics we will observe a signal at approximately the @xmath22 level .
we measure a preliminary invariant multiplicity for @xmath3 in the kinematic range @xmath23 of @xmath24 @xmath25 .
we can then extract a total penalty factor for adding a unit of strangeness in this kinematic range by taking the ratio of the @xmath3 invariant yield to that of @xmath6 in the same kinematic region .
this gives a penalty factor of @xmath26 @xmath27 .
some of this penalty is simply due to the fact that it is harder to get the all ingredients of a @xmath3 close enough in momentum space to coalesce than is the case for @xmath6 .
we can remove this part of the penalty factor by normalizing each of these yields to the product of the invariant yields of their ingredient baryons in the relevant kinematic ranges . using results from e891 for the @xmath28 yields and e864 measurements for the remaining quantities ,
we obtain @xmath29 as a preliminary result .
this then represents a penalty of approximately 6 for coalescing a strange as opposed to a non - strange state ; we note that this was predicted to be equal to 1 @xcite .
by our measurements of light nuclei with @xmath5 we can study trends in light nuclei production as a function of kinematics and centrality . in 10% central collisions in a limited kinematic range
, we can explore dependences of production on mass , spin , isospin , strangeness , and potentially binding energy of the state up to mass number @xmath30 .
an interestingly low excited state temperature of less than 15 mev is implied by our upper limit on production of @xmath19 , and our measurement of @xmath3 implies a larger than expected strangeness penalty factor .
9 a. polleri et .
al . , phys .
b. , 419 ( 1998 ) 19 .
george , proceedings of park city conference , january 1999 .
proceedings of aps heavy - ion symposium , 1999 .
world scientific .
proceedings of aps heavy - ion symposium , 1999 . world scientific . v. serfling et .
al . , phys . rev .
lett . , 80 ( 1998 ) 3928 .
a. j. baltz et .
al . , phys .
b. , 382 ( 1996 ) 35 . | we present measurements of the production of stable light nuclei for mass number @xmath0 , of strongly decaying states @xmath1 and @xmath2 and of the the hypernucleus @xmath3 .
we also examine trends in the production of these multibaryon states as a function of kinematic variables and properties of the states including strangeness content . |
ic 2554 is a southern galaxy with a peculiar optical appearance .
it is also known as se 72/1 ( srsic 1974 ) , eso 092-ig012 , and iras 100756647 .
1 shows a deep optical image of ic 2554 ( laustsen , madsen & west 1987 ) revealing many bright knots as well as a network of dust lanes intersecting the main body of the galaxy .
the optical extent of ic 2554 is 325 ( see table 1 ) .
its irregular and asymmetric appearance has prompted the division into multiple components , roughly aligned north - south .
lauberts et al .
( 1978 ) describe ic 2554 as a ` fox - shaped ' system with a complex spiral pattern .
they distinguish three components : two central condensations separated by 18 with optical velocities of 1230 and 1310 kms@xmath6 and a detached arm or satellite to the north ( 1190 kms@xmath6 ) .
corwin et al . ( 1985 ) describe ic 2554 as a colliding pair ( ic 2554a and b , 08 apart ) with plumes .
laustsen et al .
( 1987 ) describe ic 2554 as a very disturbed pair of galaxies with a large number of emission nebulae , indicating high activity of star formation .
lauberts ( 1982 ) and laustsen et al .
( 1987 ) note ic 2554 as possibly disturbed by the elliptical galaxy ngc 3136b ( eso 092-g013 ) . but
ic 2554 has also been classified as a peculiar barred spiral galaxy ( de vaucouleurs et al .
1991 ) , where the northern spiral arm appears elongated .
high - resolution hi observations ( presented here ) are needed to study the gas dynamics of this complex system and distinguish between the numerous scenarios .
close to ic 2554 ( see fig . 2 ) are three known galaxies : the elliptical galaxy ngc 3136b ( 1780 kms@xmath6 ) , the faint spiral eso 092-g009 and the galaxy rkk 1959 . for a summary of various optical properties of ic 2554 and
the neighbouring galaxies see table 1 . | atca hi and radio continuum observations of the peculiar southern galaxy ic 2554 and its surroundings reveal typical signatures of an interacting galaxy group .
we detected a large hi cloud between ic 2554 and the elliptical galaxy ngc 3136b .
the gas dynamics in ic 2554 itself , which is sometimes described as a colliding pair , are surprisingly regular , whereas ngc 3136b was not detected . the hi cloud , which emerges from ic 2554 as a large arc - shaped plume , has a size of @xmath0 kpc , larger than that of ic 2554 .
the total hi mass of the ic 2554 system is @xmath1 m@xmath2 , a third of which resides in the hi cloud .
it is possible that tidal interaction between ic 2554 and ngc 3136b caused this spectacular hi cloud , but the possibility of ic 2554 being a merger remnant is also discussed .
we also detected hi gas in the nearby galaxies eso 092-g009 and rkk 1959 as well as an associated hi cloud , atca j10066710 .
together they have an hi mass of @xmath3 m@xmath2 .
another new hi source , atca j10076659 , with an hi mass of only @xmath4 m@xmath2 was detected roughly between ic 2554 and eso 092-g009 and corresponds to a face - on low surface brightness dwarf galaxy .
star formation is evident only in the galaxy ic 2554 with a rate of @xmath54 m@xmath2yr@xmath6 .
galaxies : individual ( ic 2554 , ngc 3136b , eso 092-g009 , rkk 1959 ) galaxies : interactions |
reactor neutrino experiments have played a critical role in the history of neutrinos . among them , savannah river experiment @xcite by reines and cowan in 1956 observed the first neutrino .
chooz @xcite determined the most stringent upper limit of the last unknown neutrino mixing angle @xmath1 in 1998 .
kamland @xcite observed the first reactor neutrino disappearance in 2003 .
flux is an important source of uncertainties for a reactor neutrino experiment .
it is determined from thermal power measurements , reactor core simulation , and knowledge of neutrino spectra of fuel isotopes .
past reactor neutrino experiments have determined the flux to ( 2 - 3)% precision . in the coming years , three precision experiments on neutrino mixing angle @xmath0
using reactor neutrinos , daya bay @xcite , double chooz @xcite , and reno @xcite , will start operation .
all these experiments use near - far detector configurations .
most uncertainties from the reactor will cancel out .
the residual error will range from 0.1% to 0.45% .
however , correlation among reactor cores need better understanding of the error sources .
recently there are increasing interests on non - proliferation monitoring @xcite using ton - level neutrino detectors .
a reactor neutrino experiment at an intermediate baseline @xmath260 km with a giant detector @xcite will have rich physics content .
precise determination of neutrino flux will greatly improve the sensitivity of these experiments . another category of reactor neutrino experiments is @xmath3-electron or @xmath3-nucleus scattering experiments , such as texono @xcite , munu @xcite , gemma @xcite , etc .
normally they wo nt rely on precise neutrino flux . in this note , we will review the calculation of the reactor neutrino flux and recent progresses on the error analysis .
most commercial reactors are pressurized water reactor ( pwr ) or boiling water reactor ( bwr ) .
they are very similar in neutrino flux calculation .
we will use pwr as examples in the following .
the @xmath4u enrichment in fresh fuel of a pwr is normally ( 3 - 4)% , and more than 95% is @xmath5u .
electron antineutrinos are emitted from subsequent @xmath6-decays of fission fragments .
they are dominated by 4 isotopes , @xmath4u , @xmath7pu , @xmath8pu , and @xmath5u .
other isotopes contribute only at 0.1% level .
one can calculate the neutrino energy spectrum of each isotope by summing all fission fragment @xmath6-decay branches .
there are a lot of efforts on such studies . however , the fission products are very complex . due to lack of accurate nuclear data ,
such calculations carry large uncertainties at the 10% level .
the most accurate neutrino spectra of the first 3 isotopes were determined at ill @xcite by measuring the @xmath6 spectra of fissioning .
the @xmath6 spectra are then converted to neutrino spectra , with an average uncertainty 1.9% .
the @xmath5u spectrum are calculated theoretically @xcite .
they are shown in fig .
[ fig : spectra ] . , scaledwidth=40.0% ]
the isotope concentration in fuel will evolve during reactor operation as @xmath4u depletes and @xmath7pu and @xmath8pu breed .
the @xmath5u concentration is relatively stable .
such evolution can be obtained by core simulation . a typical isotope evolution as a function of operation time , in terms of fission rates of the reactor , is shown in fig . [ fig : evol ] .
, scaledwidth=40.0% ] when we know the fission rates of each isotope from core simulation , and neutrino energy spectrum of each isotope , we can easily get the neutrino flux @xmath9 , where @xmath10 is the fission rate of isotope @xmath11 and @xmath12 is its neutrino spectrum . however , the fission rates are proportional to the thermal power of the core , which is fluctuating .
it is unrealistic to repeat core simulation to reflect the power fluctuation .
normally we scale the neutrino flux to the measured thermal power .
@xmath13 where @xmath14 is the thermal power , @xmath15 is the energy release per fission for isotope @xmath11 , and @xmath16 is the sum of @xmath10 , thus @xmath17 is the fission fraction of each isotope . among the inputs ,
the thermal power data is provided by the nuclear power plant . the uncertainty is generally estimated to be ( 0.6 - 0.7)% @xcite .
fission fractions are obtained by core simulation as a function of burn - up .
burn - up is the amount of energy in mega watt days ( mwd ) released per unit initial mass ( ton ) of uranium ( tu ) .
the simulated fission fraction carries @xmath25% uncertainties from statistics of hundreds of analyses for various codes and various reactors @xcite .
the 5% fission fraction uncertainties corresponds to @xmath20.5% uncertainty in neutrino yield .
energy release per fission varies slightly for different cores at different time due to neutron capture and non - equilibrium products .
average numbers @xcite can be used , with uncertainties of ( 0.30 - 0.47)% .
alternatively we can extract them from the core simulation to accurately reflect the core differences and burn - up effects .
recently there are studies to include contributions from non - equilibrium isotopes in the core @xcite as well as that from spent fuel which is temporarily stored adjacent to the core @xcite .
these are at sub - percent level and only contribute to the low energy region .
besides fission products , @xmath5u(n,@xmath18)@xmath7u reaction also contributes to the neutrino yield .
it is below inverse @xmath6-decay threshold ( 1.8 mev ) but will contribute significantly to low energy @xmath3-electron scattering experiments @xcite .
the most accurate thermal power measurement is the secondary heat balance method .
detailed description of this measurement can be found , for example , in @xcite .
this is an offline measurement , normally done weekly or monthly .
the uncertainty is cited as 0.7% by chooz and palo verde .
primary heat balance tests are online thermal power measurement .
normally it is calibrated to the secondary heat balance measurement weekly .
daya bay power plants control the difference of these two measurements to less than 0.1% of the full power .
these data are good for neutrino flux analysis .
to 0.1% level , it can be taken as the secondary heat balance measurement .
the power plants also monitor the ex - core neutron flux , which gives the nuclear power . this monitoring is online , for safety and reactor operation control .
it is normally calibrated to the primary heat balance measurement daily .
this measurement is less accurate , controlled to be less than 1.5% of the full power by daya bay power plant .
recently there are a lot of studies on the power uncertainties and instrumentation improvements by the power plants @xcite , with the motivation of power uprates .
the power measurements can be more accurate than what were cited in the past reactor neutrino experiments .
the uncertainties of the secondary heat balance is dominated by the flow rate measurement . in the past , there are two kinds of widely used flow meters , venturi type flow meters and orifice plate flow meters .
venturi flow meters are used by most us and japan reactors .
the uncertainty is often 1.4%
. it can be as low as 0.7% if properly calibrated and maintained .
but they suffer from fouling effects , which could grow as high as 3% in a few years .
to improve the measurement , ultrasonic flow meters have begun to be in use in some us and japan reactors .
they have uncertainties 0.45% for type i and 0.2% for type ii @xcite .
the orifice plate flow meters are used by french reactors .
they have no fouling effects .
typically they have an uncertainty of 0.72% and could be improved to 0.4% with laboratory tests .
it should be noted that the uncertainties of above flow meters are at the 95% c.l .
( confidence level ) , as defined in iso-5167 .
unless specified , the thermal power uncertainty given by the power plant is also at 95% c.l .
an edf ( electricite de france ) n4 reactor with four parallel steam generators , which is the chooz type , is analyzed in @xcite .
main components of the uncertainties are shown in table [ tab : n4 ] .
it is dominated by the discharge coefficient , which is an empirical formula in the flow rate measurement and its uncertainty is specified in iso 5167 - 1 - 2003 .
the final uncertainty of the thermal power is 0.40% at 95% c.l . in this evaluation
, it is assumed that the discharge coefficients of the orifice plates in four coolant loops are independent , thus the final uncertainty is statistically reduced .
if the discharge coefficients are fully correlated for all four orifice plates , then there is no statistical reduction .
the power uncertainty will be 0.37% at 1@xmath19 level , which is still significantly smaller than 0.7% that chooz used . .error table for the thermal power measurements for n4 reactor @xcite.[tab : n4 ] [ cols="^,^,^ " , ]
electron antineutrinos are emitted from the subsequent @xmath6-decays of fission fragments . due to
the lack of data for the @xmath6-decays of the complex fission products , theoretical calculations of the neutrino spectra of isotopes carry large uncertainties .
ill @xcite measured the @xmath6 spectra of fissioning of @xmath4u , @xmath7pu , and @xmath8pu by thermal neutrons , and converted them to neutrino spectra .
the normalization error is estimated to be 1.9% .
spectrum shape error is from 1.34% at 3 mev to 9.2% at 8 mev , as shown in fig .
[ fig : spectra ] .
@xmath5u can not fission with thermal neutrons .
its spectrum relies on theoretical calculation .
the uncertainty is estimated to be 10% @xcite .
normally @xmath5u contributes ( 7 - 10)% of fissions in a pwr .
the calculated neutrino counting rate and spectra were verified by bugey and bugey-3 @xcite .
the normalization error is further lowered to 1.6% , which was used by chooz .
the ill spectra are derived after 1.5 days exposure time with thermal neutron .
thus , long - lived fission fragments have not reached equilibrium . in a real reactor
, these fission products will accumulate and contribute to the neutrino flux .
chooz estimated this contribution to be @xmath20.3% on the average and ignored it in the detailed analysis due to its small size comparing to other errors .
six chains have been identified in @xcite , with half lives from 10 hours to 28 years .
they only contribute to the low energy region .
further studies show that for a typical pwr , on average these contributions are @xmath20.2% of total neutrino detection rate via inverse @xmath6-decay @xcite . in the 2 - 4 mev region
, it increases to 0.8% after one year s accumulation , as shown in fig .
[ fig : noneq ] -decay cross section @xcite.[fig : noneq ] , scaledwidth=40.0% ] spent fuel is normally stored temporarily adjacent to the core .
the storage could be as long as 10 years .
similar to the non - equilibrium contributions , the long - lived fission fragments in the spent fuel will contribute to the neutrino flux .
a pwr is normally refueled very 12 - 18 months .
the spent fuel from one refueling will contribute @xmath20.2% of the total neutrino rate after first several days .
before the 1980 s , the reactor neutrino flux was determined to an uncertainty of 10% . with a lot of efforts , especially by ill , bugey , chooz , palo verde , etc .
, the uncertainty has improved to ( 2 - 3)% .
motivated by high precision neutrino measurements by scientific collaborations and power uprates by the power plants , we have more accurate thermal power .
the uncertainty could be lowered from 0.7% to 0.4% .
small corrections from non - equilibrium isotopes and spent fuel , as well as energy release per fission are studied in detail .
we also have a global picture of uncertainties of fission rate simulations .
however , there is no new data on neutrino spectra of fuel isotopes . for single - detector experiments , the neutrino spectrum uncertainty of about 2% will dominate .
the next @xmath0 experiments with near - far relative measurements will suffer little from reactor flux uncertainties , which is estimated to be 0.1% to 0.45% , depending on the experiment layout .
the correlation among reactor uncertainties is important for @xmath0 experiments since correlated errors will cancel out .
meanwhile , high precision detector and high statistics at the near detector of these experiments may help to improve the knowledge of neutrino spectra .
since the time that this manuscript was written , it has been suggested @xcite that the antineutrino spectra for all of the relevant fission isotopes ( @xmath4u , @xmath5u , @xmath7pu , and @xmath8pu ) are several percent larger than those of ref
. @xcite .
this issue requires further attention . | flux is an important source of uncertainties for a reactor neutrino experiment .
it is determined from thermal power measurements , reactor core simulation , and knowledge of neutrino spectra of fuel isotopes .
past reactor neutrino experiments have determined the flux to ( 2 - 3)% precision .
precision measurements of mixing angle @xmath0 by reactor neutrino experiments in the coming years will use near - far detector configurations .
most uncertainties from reactor will be canceled out .
understanding of the correlation of uncertainties is required for @xmath0 experiments .
precise determination of reactor neutrino flux will also improve the sensitivity of the non - proliferation monitoring and future reactor experiments .
we will discuss the flux calculation and recent progresses .
neutrino flux , reactor , fission rate |
in a close binary consisting of a stellar - mass black hole and a mass donating normal star , the accreting matter releases its gravitational energy as x - ray radiation . in fig . [
fig : cygx1 ] , we show two typical x - ray spectra of cyg x-1 , the best studied black hole binary , showing its two distinct spectral states . in the low / hard state ,
its spectrum is approximately a power - law , with an exponential cut - off near @xmath3 kev .
by contrast , the high / soft state is characterized by dominant soft emission below 10 kev , accompanied by a hard power - law tail extending to 500 kev and beyond .
as described , e.g. , by makishima et al .
( 1986 ) , the soft spectral component is interpreted as thermal emission from an optically thick accretion disk around the black hole , as it can be well reproduced by a multi - color disk model ( mcd model ; mitsuda et al .
this model approximates a spectrum from a standard accretion disk ( shakura & sunyaev 1973 ) . under this standard disk solution
, the keplerian rotational energy takes up half of the gravitational energy , while the remaining half is radiated as approximately blackbody emission from the disk surface .
the mcd model ignores any inner disc boundary condition , describing the disk local temperature @xmath4 as @xmath5 where @xmath6 and @xmath7 are the two spectral parameters of the mcd model , the maximum observed disk temperature and the apparent disk inner radius , respectively .
the latter is related to the true inner radius , @xmath8 by @xmath9 where @xmath10 is a correction to account for a stress free inner boundary condition ( kubota et al .
1998 ) and @xmath112.0 is a colour temperature correction factor to account for distortions of the spectrum from a true blackbody ( shimura & takahara 1995 ) .
the disk bolometric luminosity @xmath0 can be related to these two spectral parameters as @xmath12 where @xmath13 is the stefan - boltzmann constant .
this model function is simple , and well reproduces the soft spectral component of high / soft state black hole binaries with physically meaningful fit parameters .
for example , the soft component of cyg x-1 shown in fig .
[ fig : cygx1 ] was well fitted with the mcd model of @xmath14 km ( dotani et al .
1997 ) with @xmath15 and @xmath10 . since the black hole mass of cyg x-1 was estimated to be 1016 @xmath16 from several optical observations ( gies & bolton 1986 , ninkov et al .
1987 , liang & nolan 1983 ) , the last keplarian orbit , @xmath17 , of a non - spinning black hole is calculated as @xmath18140 km for cyg x-1 .
therefore , it is found that @xmath8 determined by the x - ray observations well agrees with @xmath17 calculated by referring to the optical observations .
even more compelling evidence for the standard disk formalism is given by the observational fact that the value of @xmath8 is usually observed to remain constant in the high / soft state as @xmath0 changes significantly .
this indicates that @xmath8 reflects the physical parameters of the black hole itself .
figure [ fig : lmcx3 ] shows the values of @xmath0 obtained by spectral fits of @xmath1980 rxte / pca observations of high / soft state spectra from lmc x-3 , a well known black hole binary system .
this figure shows that all the data points lie along the solid line which represents @xmath20 ( requiring constant emitting area i.e. constant @xmath7 , and constant colour temperature correction , @xmath21 , in equation [ eq : ldisk ] ) over the range of @xmath228@xmath23(@xmath191080% of the eddington limit , @xmath24 , of a @xmath25 black hole .
this relation of @xmath20 has been widely observed in many high / soft - state black hole binaries except at very high disk temperature ( typically at @xmath261.2 kev , see below ) .
( e.g. , tanaka & lewin 1995 ; mcclintock & remillard 2003 ; ebisawa et al .
1994 ; gierlinski & done 2004 ) . despite these sucesses
, standard disc models strongly conflict with the observed disc emission on several points .
( 1 ) the @xmath27 viscosity parameterisation of the standard shakura - sunyaev disc model is unstable in the range @xmath19 0.11 @xmath24 where radiation pressure is dominant but the disc is not yet optically thick enough to regain stability through advecting the radiation ( slim disk solution ; abramowicz et al .
yet the observed disk emission is very stable in this luminosity range .
( 2 ) theoretical models of radiation transfer through the disc vertical predict that the colour temperature correction factor should change with luminosity . however , the amount of this change has now been shown to be rather small when true metal opacities are included ( davis et al 2005 ) .
this superceeds previous work which included only h and he , where the calculations of merloni et al ( 2000 ) predicting a large change in @xmath21 , while those of shimura & takahara ( 1995 ) had @xmath21 varying by less than 20% .
the resolution of this discrepancy may be that merloni et al ( 2000 ) assumed a vertical density profile which is appropriate for a radiation pressure dominated disc , but their lowest luminosity discs are gas pressure dominated ( gierlinski & done 2004 ) . thus while the second point is now not necessarily in conflict with the standard disk models , the stability of the observed high / soft state spectra can give constraints on theoretical models of accretion disc structures .
however , high @xmath28 spectra are not always disk dominanted .
there is an alternative bright spectral state , termed `` very high '' or `` steep power law '' state which was discovered by miyamoto et al .
( 1991 ) and is characterised by both strong disk _ and _ strong ( and rather steep ) power law emission which is highly variable ( e.g. van der klis 1994 ; mcclintock & remillard 2004 ) .
when the spectra in this state are fitted with the canonical mcd plus power - law model , the data no longer show @xmath29 , implying that the estimated values of @xmath8 ( and/or the colour temperature correction , @xmath21 ) are no longer constant .
this anomalous behavior is clearly seen in fig .
[ fig : hr ] which shows results of the mcd plus power - law fit to the pca data of two high state black hole binaries , gro j@xmath30 and xte j@xmath31 ( kubota , makishima , ebisawa 2001 , kme01 ; kubota & makishima 2004 , km04 ) .
the data points indicated with open circles show significant variation in both @xmath6 and @xmath8 , while those with filled circles show constant values of @xmath8 .
this anomaly was found when the spectra have a strong power law tail , carrying more than @xmath19 30 per cent of the luminosity ( corresponding to a hardness ratio larger than @xmath19 0.60.7 shown in the top panels of fig .
[ fig : hr ] ) . another breakdown in the @xmath32 relation can also be seen in fig .
[ fig : hr ] for highly luminous _
disk dominated _ spectra with negligible power - law tail ( filled squares : details are in kme01 and km04 ) .
however , the obtained values of @xmath8 are slightly ( but systematically ; @xmath191525% in gro j@xmath30 and @xmath191015% in xte j@xmath31 ) smaller than those in the typical high / soft state indicated with filled circles in fig .
[ fig : hr ] .
these violations can be characterized in fig .
[ fig : tl1 ] in which @xmath0 is plotted against @xmath6 . in this figure ,
data points with filled circle are consistent with solid lines showing constant @xmath8 via equation ( [ eq : ldisk ] ) , and thus they are in good agreement with the standard picture as is the case of lmc x-3 .
we thus called these data points standard regime in kme01 and km04 . on the other hand
, the data points with open circles , show significant deviation from the solid lines in both gro j@xmath30 and xte j@xmath31 , so were termed anomolous regime by kme01 and km04 , or `` weak very high state '' by kubota & done ( 2004 ) .
the data points with filled squares show the highest @xmath0 .
their deviation is much milder than that of the anomalous regime , and their spectral shape is similar to that of the standard regime .
this was termed `` apparently standard regime '' by km04 .
figure [ fig : spec ] shows examples of typical pca spectra of xte j@xmath31 in each regime .
( a)gro j165540 + ( b)xte j1550 - 564 + ( a)gro j165540 + ( b)xte j1550 - 564 + ( a)gro j165540 + ( b)xte j1550 - 564 +
a clue to the problem in the anomalous regime was given by kme01 and km04 .
they showed that the anomolous regime was seen when there was enhanced hard x - ray power law emission compared to the disc dominated spectra .
this shows that there is significant inverse compton scattering of the disk photons by some high energy electrons , probably in a corona over the disk .
the spectral shape requires that the compton region is mildly optically thick , so the disk luminosity is artificially suppressed by @xmath33 . in fig .
[ fig : tl2 ] we approximately correct for this by plotting @xmath34 ( @xmath35 is the bolometric luminosity of the thermal compton component ) instead of @xmath0 , against @xmath6 based on this spectral fits .
we found that the anomolous regime points in fig .
[ fig : tl1 ] settled back to the solid line , consistent with a constant inner radius disk in this and the disk dominated states .
other approximations for the correction to the observed disk luminosity ( e.g. assuming that compton scattering conserves photon number ) give similar results ( km04 ) .
a mildly optically thick , thermal inverse compton scattering is generically required in all broad bandpass data of the very high state ( gs 2000 + 25 , gs 1124 - 68 : zycki , done & smith 1999 ; xte j1915 + 105 : zdziarski et al 2002 ; xte j1550564 , gierlinski & done 2003 ; gro j@xmath30 : kobayashi et al .
2003 ) .
it is interesting to note here that qpos are preferentially observed in the anomalous regime in both gro j@xmath30 ( remillard et al .
1999 ) and xte j@xmath31 ( remillard et al . 2002 ) , and the energy spectra of the qpo are also follow the spectrum of the comptonised emission . therefore , the qpos should have some relation to the existence of the compton cloud .
the apparently standard regime corresponds to the most luminous phase of the outburst , and the data in this regime occupies the upper - right region of the @xmath0-@xmath6 diagram ( fig .
[ fig : tl2 ] ) . as seen in fig .
[ fig : spec ] , the spectra in the apparently standard regime show a dominant soft component accompanied by a very weak hard tail .
although these properties are similar to the standard regime , the apparently standard regime is slightly different from the standard regime because the disk luminosity does not increase with @xmath36 , but rather more slowly , more like @xmath37 .
thus either the disk inner radius is decreasing , or the colour temperature correction is increasing ( or both ) .
moreover , km04 showed that the canonical spectral model often failed to give acceptable fits to the data ( fig .
[ fig : spec ] ; see also fig . 2 in km04 ) . in order to quantify the difference , km04 constructed a generalization of the mcd model in which the radial temperature dependance is @xmath38 .
this @xmath39-free disk model differs from the standard mcd model which has @xmath39 fixed at @xmath40 , so there are now three fit parameters , @xmath6 , @xmath7 , and @xmath39 ( see also hirano et al.1995 , mineshige et al .
1994 ) . in fig .
[ fig : t - p ] , the best fit values of @xmath39 are plotted against @xmath6 for lmc x-3 , gro j@xmath30 and xte j@xmath31 . though the absolute values of @xmath39 were slightly affected by fitting conditions ( whether @xmath41 or @xmath42 were fixed or free , including ( or excluding ) a gaussian for a narrow iron line ) , the characteristic @xmath39-@xmath6 behavior on fig .
[ fig : t - p ] was not affected by them . as a calibration trace on @xmath39-@xmath6 plane for the standard regime ,
we show a result of lmc x-3 in fig .
[ fig : t - p]a .
this source stays in the standard regime up to @xmath43 , and shows positive correlation between @xmath39 and @xmath6 . with the pca data , this positive correlation is considered as an artifact which is present even if the standard accretion disk is realized .
this artifact happens because the mcd model ignores the boundary condition at the inner edge of the standard disk solution .
the actual temperature gradient of a standard accretion disk must be flatter than 3/4 near the innermost disk edge , where the temperature will approach zero .
as @xmath6 decreases , the limited pca band pass will sample preferentially the emission from inner disk regions , thus making @xmath39 appear smaller than 3/4 as is found in lmc x-3 .
as for xte j@xmath31 , while the data points for @xmath44kev ( standard regime ) are well consistent with the calibrated trace in fig .
[ fig : t - p]c , those for @xmath45 kev ( in the apparently standard regime ) appear well below the trace ( and in fact , @xmath39 _ negatively _ correlates to @xmath6 ) .
similarly , the data points of gro j@xmath30 for @xmath461.1 kev are mostly consistent with the trace , but those for @xmath47 kev are below the trace .
therefore , the observed temperature gradient in the apparently standard regime is smaller than in the standard regime .
this result means that the radiative efficiency at the inner portion of the accretion disk in the apparently standard regime is lower than that in the standard regime ( case of the standard disk ) .
hence the accretion disk in the apparently standard regime deviates from the standard disk and requires another cooling process in addition to radiative cooling .
the advection in the optically thick disk is a candidate for the cooling process , and in fact , the slim disk solution , seen in a top slope of the @xmath48-shape sequence ( abramowicz et al .
1995 ) , predicts smaller values of @xmath39 than the standard disk ( e.g. , watarai et al .
2000 ) .
we show that there are three spectral regimes in the stable high / soft state black hole binaries , the standard regime , the anomalous ( weak very high ) regime , and the apparently standard regime .
the last two regimes appear typically when the x - ray luminosity exceeds a certain critical luminosity .
the accretion disk structures are somewhat different in these three regimes , but all are consistent with the disk extending down to a constant inner radium , most probably the innermost stable keplerian orbit around the central black hole .
the accretion disk in each regime is characterized by a radiation dominated standard disk but in the anomolous regime there is also high energy plasma which causes the observed strong compton scattering , while in the apparently standard regime there is some change to the cooling process , perhaps marking the onset of advection cooling in addition to radiative cooling .
this picture can also explain many other black hole binaries in the high accretion rate state including 4u @xmath49 ( abe et al .
based on their x - ray luminosities , ultraluminous compact x - ray sources ( ulxs ) in nearby galaxies are supposed to be intermediate mass ( @xmath1930100 @xmath16 ) black holes ( makishima et al .
2000 ) , super eddington stellar mass black holes , or ( mild ) beaming black holes ( e.g. , king et al .
2001 ) . among many ulxs ,
mizuno et al . ( 2001 ) analyzed three bright ( 0.21.5@xmath50 ) variable ulxs observed with asca , ic 342 source 1 , m81 x-6 ( sometimes called x-11 ) and ngc 1313 source b. they found that the x - ray spectra are well reproduced by the mcd model and that obtained values of @xmath8 are not constant but variable as @xmath51 .
this is very similar to that of xte j@xmath31 in the apparently standard regime .
in addition , kubota et al .
( 2002 ) analyzed the other spectral data of ic 342 source 1 obtained with asca in different date . during that observation ,
the source showed a power - law like spectrum rather than the mcd shape , and it was reanalyzed in the framework of the inverse comptonization as is the case of the anomalous regime .
they found that this power - law like spectrum then lies nicely on the same luminosity - temperature relation as defined by the mcd - type spectra given by mizuno et al .
( 2001 ) .
we can see that the data points of bright ulxs on @xmath6-@xmath0 diagram is understood by shifting up those of stellar black hole , xte j@xmath31 in the brightest regimes ( see fig . 2 by kubota et al .
these spectral studies thus supports the scenario of high accreting relatively massive ( more massive than galactic stellar black holes ) black holes for ulxs .
999 abe , y. , fukazawa , y. , kubota , a. , et al .
pasj in press abramowicz , m. a. , czerny , b. , lasota , p. , et al .
apj , 332 , 646 - 658 , 1988 abramowicz , m. a. , chen , x. , kato , s. , et al .
apjl , 438 , 37 - 39 , 1995 balucinska , m. , & hasinger , g. , a&a 241 , 439 - 450 , 1991 cui , w. , ebisawa , k. , dotani , t. , et al .
apj 493 , l75 - 78 , 1998 davus , s. w. , blaes , o. m. , hubeny , i. , et al .
apj 621 , 372 - 387 , 2005 dotani , t. , inoue , h. , mitsuda , k. , et al .
apj 485 , l87 - 90 , 1997 ebisawa , k. , ogawa , m. , aoki , t. , et al . , pasj , 46 , 375 - 394 , 1994 gierlinski , m. , done , c. , mnras 347 , 885 - 894 , 2004 gierlinski , m. , done , c. , mnras 342 , 1083 - 1092 , 2003 gierlinski , m. , zdziarski , a.a . , done , c. , et al . ,
mnras 288 , 958 - 964 , 1997 gies , d.r . , bolton , c.t .
, apj 304 , 371 - 393 , 1986 grove , j.e . , johnson , w.n . ,
kroeger , r.a . , et al . ,
apj , 500 , 899 , 1998 hirano , a. , kitamoto , s. , yamada , t. , et al .
apj , 446 , 350 , 1995 king , a.r . , davies , m. b. , ward , m.j . ,
et al . , apj , 552 , 109 - 112 , 2001 kobayashi , y. , kubota , a. , nakazawa , et al .
, pasj , 55 , 273 - 279 , 2003 kubota , a. , done , c .
, mnras , 353 , 980 - 990 , 2004 kubota , a. , done , c. , makishima , k. , mnras 337 , l11-l15 , 2002 kubota , a. , makishima , k. , apj 601 , 428 - 438 , 2004 ( km04 ) kubota , a. , makishima , k. , ebisawa , k. , apjl , 560 , 147 - 150 , 2001 ( kme01 ) kubota , a. , makishima , k. , ueda , y. , et al . ,
pasj , 50 , 667 - 673 , 1998 liang , e.p . , nolan , p.l . , ssrv 38 , 353 - 384 , 1984 mcclintock , j.e . ,
& remillard , r. a. , in compact stellar x - ray sources , eds .
lewin , w.h.g .
, and van der klis , m. , cambridge university press , preprint ( astro - ph/0306213 ) 2003 mcconnell , m. , forrest , d. , ryan , j. , et al .
apj 424 , 933 - 939 , 1994 makishima , k. , kubota , a. , mizuno , t. , et al .
apj 535 , 632 - 643 , 2000 makishima , k. , maejima , y. , mitsuda , k. , et a.l . , apj 308 , 635 - 643 , 1986 merloni , a. , fabian , a. c. , ross , r.r . , mnras 313 , 193 - 197 , 2000 mineshige , s. , hirano , a. , kitamoto , s. , et al .
apj , 426 , 308 - 312 , 1994 mitsuda , k. , inoue , h. , koyama , k. , et al . pasj , 36 , 741 - 749 , 1984 miyamoto , s. , kimura , k. , kitamoto , s. et al . , apj , 383 , 784 - 807 , 1991 mizuno , t. , kubota , a. , & makishima , k. , apj 554 , 1282 - 1289 , 2001 ninkov , z. , walker , g.a.h . , yang , s. , apj 321 , 425 - 437 , 1987 phlips , b. f. , jung , g.v .
, leising , m.d . , et al . ,
a&as 120 , 155 - 158 , 1996 remillard , r.a . ,
morgan , e.h .
, mcclintock , j.e . , et al . ,
apj , 522 , 397 - 412 , 1999 remillard , r.a . , sobczak , g.j . ,
muno , m.p . ,
mcclintock , j.e . , apj , 564 , 962 - 973 , 2002 shimura , t. & takahara , f. apj , 445,780 - 788 , 1995 shakura , n.i . ,
sunyaev , r. a. , a&a , 24 , 337 - 355 , 1973 tanaka , y. , & lewin , w. h. g. in x - ray binaries , eds . w. h. g. lewin , j. van paradijs , and w. p. j. van den heuvel ( cambridge university press , cambridge ) , pp.126 - 174 , 1995 van der klis , m. , in x - ray binaries eds . w. h. g. lewin , j. van paradijs , and w. p. j. van den heuvel ( cambridge university press , cambridge ) , pp.252 - 307 , 1995 watarai , k. , fukue , j. , takeuchi , m. , et al.,pasj , 52 , 133 , 2000 zdziarski , a. a. , grove , j. e. poutanen , j. , rao , a. r. , vadawale , s. v. apj , 554 , l45-l48 , 2001 zycki , p. t. , done , c. , smith , d. a. mnras , 305 , 231 - 240 , 1999 | we outline a framework for understanding the x - ray spectra of high mass accretion rate stellar black holes based on x - ray data from rxte and asca .
three spectral regimes can be separated out by the behaviour of the observed disk luminosity and temperature .
the well established `` standard regime '' is seen when the disk dominates the spectrum , where only a small fraction of the luminosity is emitted in the power law tail .
these spectra generally satisfy the standard relation expected for thermal emission from a constant area , namely that the disk bolometric luminosity , @xmath0 , is proportional to its maximum temperature , @xmath1
. however , at higher luminosities this starts to change to @xmath2 .
this `` apparently standard regime '' is still dominated by the disk emission , but this difference luminosity - temperature relation and subtle changes in spectral shape may show that another cooling process is required in addition to radiative cooling . at intermediate luminosities
there is an anomalous regime ( or weak very high state ) where the disk temperature and luminosity are less clearly related .
these spectra are characterized by the presence of a much stronger comptonized tail indicating high energy electrons .
when observed disk emission is corrected for the the effects of comptonisation then these points lie back on the standard relation .
the growth of this comptonising corona is also clearly linked to the quasi - periodic oscillations , as these are observed preferentially in the anomalous regime .
this presented picture was found to explain the spectral behavior of both black hole binaries in our galaxy and lmc .
spectral evolution of several bright ulxs observed with asca were also successfully explained in the same picture . , black holes ; accretion disk . |
supermanifolds were introduced by berezin , leites and konstant in the 1970s as a mathematical framework for the quantum theory of commuting and anticommuting fields .
a remarkable contribution was berezin s definition of his integral , in ref .
@xcite , predating the definition of supermanifolds by several years , and providing at the time sufficient indication that a reasonable supersymmetric analysis should exist . despite its utility ,
the integral suffers from a fundamental pathology : only the integral of _ compactly supported _ quantities is well - defined in a coordinate independent form
changes of variables introduce , in general , so - called _ boundary terms_. this can be seen as a major obstacle in the development of global superanalysis .
for example , although stokes s theorem @xmath0 has been extended to supermanifolds by bernstein and leites @xcite , this extension supposes that the supermanifold structure on the boundary @xmath1 enjoys a rather strong compatibility requirement .
in fact , even for compactly supported integrands @xmath2 , the conclusion of the theorem _ fails _ in general , unless this assumption is made ( _ cf . _ below ) .
an invariant definition of the integral can however be made , on the basis of the following simple observation : for any supermanifold @xmath3 , there exist morphisms @xmath4which we call _ retractions_which are left inverse to the canonical embedding @xmath5 .
any retraction @xmath6 is a submersion whose fibres have compact base ; thus , there is a well - defined _ fibre integral _
@xmath7 which takes berezin forms on @xmath3 to volume forms on @xmath8 , and one may define @xmath9 taking pullback retractions , this definition is now trivially well - defined under coordinate changes . furthermore , whereas retractions are non - unique in general , for certain classes of supermanifolds_e.g .
_ , lie supergroups @xmath10 , homogeneous @xmath10-supermanifolds , and superdomains there exist canonical retractions .
this framework allows us to give an explicit description of the behaviour of the integral under coordinate changes .
to state our main result ( ) , let @xmath11 be an open subspace of a supermanifold whose underlying space @xmath12 is a manifold with corners .
that is , we have @xmath13 for some functions @xmath14 which define _ boundary manifolds _ @xmath15 .
let @xmath6 , @xmath16 be retractions on @xmath17 . on each @xmath18 ,
one considers the supermanifold structure @xmath19 induced by @xmath20 and the retraction @xmath21 induced by @xmath6 .
let @xmath22 be even vector fields such that @xmath23 on suitable neighbourhoods of @xmath24 .
then , for any berezin density @xmath2 such that the integrals exist , @xmath25 here ,
we sum over all @xmath26 and all multi - indices @xmath27 ; moreover , @xmath28 and @xmath29 denotes the multi - index @xmath30 with entries reduced by one .
the differential operators on the right hand side are of degree up to @xmath31 . from this change of variables formula
, we deduce a version of stokes s theorem which is valid for an arbitrary supermanifold structure on the boundary ( ) . compared to equation , the right hand side depends not only on @xmath32 , but on transversal derivatives up to order @xmath31 .
the question of defining the integral of non - compactly supported berezinians was first studied by rothstein @xcite in his seminal paper .
his fundamental insight was that the integral becomes well - defined if instead of the berezinian sheaf , one considers the sheaf of super - differential operators with values in volume forms . this insight is vital indeed , rothstein s techniques form the basis of our investigations , and one may view equation as an attempt to translate rothstein s definition of the berezin integral _ via _ the ` fermi integral ' to the realm of ordinary berezinians . for applications to superanalysis ,
rothstein s sheaf is somewhat unwieldy , since it is an @xmath33-module of infinite rank .
for example , in the context of homogeneous supermanifolds , one frequently fixes integrands by invariance .
of course , this can only be done for @xmath33-modules of rank one , which favours the berezinian sheaf as a tool for superanalysis .
the applications we have in mind come from the spherical harmonic analysis on riemannian symmetric supermanifolds , in particular , the study of orbital and eisenstein integrals in the spirit of harish - chandra . besides its relation to representation theory @xcite ,
this subject is of high current interest in mathematical physics , in the study of @xmath34-model approximations of invariant random matrix ensembles , as are applied to disordered metals and topological insulators @xcite .
let us end with a brief synopsis of our paper .
in , we recall some basic facts and define the integral of berezin densities with respect to a retraction . in
, we prove a version of stokes s theorem in this setting ( ) . here , the supermanifold structure on the boundary has to be chosen compatibly ( see below ) . in , we prove a version of our change of variables formula in terms of coordinates ( ) . here , the ` boundary ' nature of the ` boundary terms ' is not yet evident .
this is finally accomplished in , where the language and technique of supermanifolds with corners and boundary supermanifolds is introduced ; here , the point of view of retractions proves particularly fruitful . by applying this machinery , we prove our main result ( ) and illustrate its use in some examples . finally , we deduce a generalised stokes s theorem ( ) where the supermanifold structure on the boundary is arbitrary .
_ acknowledgements_. the present paper is based on the diploma thesis @xcite of w.p . under the guidance of j.h .
we use the standard definition of supermanifolds in terms of ringed spaces . for basic facts on these ,
we refer the reader to @xcite .
let us fix our notation .
given an object in the graded category , we will denote the underlying ungraded object by a subscript @xmath35 .
we denote supermanifolds as @xmath36 , @xmath37 , _ etc . _
unless the contrary is stated explicitly , we will assume @xmath3 , @xmath17 to be of dimension @xmath38 .
manifolds will always be hausdorff and second countable . by writing @xmath39
we will mean that @xmath40 is the ringed subspace @xmath41 of @xmath3 given by the open subset @xmath42 .
thus , unions and finite intersections of open subspaces are defined .
further , the set of superfunctions @xmath43 on @xmath40 is abbreviated by @xmath44 .
morphisms of supermanifolds @xmath45 are denoted @xmath46 , with underlying smooth map @xmath47 and the sheaf morphism @xmath48 . for a given supermanifold @xmath3
we denote the canonical embedding by @xmath5 . given @xmath49 , we write @xmath50 for @xmath51 .
now we introduce a certain type of morphisms which will be central for the following developments .
a morphism @xmath4 is called a _ retraction _ if it is a right inverse of the canonical embedding @xmath52 , _
@xmath53 in the literature , the subalgebra @xmath54 is called a _ function
factor_. it is a known fact that retractions always exist on ( real ) supermanifolds ( * ? ? ?
* lemma 3.2 ) . however , they are in general not unique . for superdomains there
exists a canonical choice of retraction . using exponential charts
, one may also give canonical retractions in the case of lie supergroups ; this can also be extended to the case of homogeneous supermanifolds .
we will repeatedly use the following standard fact ( * ? ? ?
* theorem 2.1.7 ) .
[ prop:3.3 ] let @xmath3 , @xmath17 be supermanifolds , @xmath55 a coordinate system on @xmath17 , and @xmath56 a family of superfunctions on @xmath3 where the @xmath57 are even and the @xmath58 are odd .
then there exists a unique morphism @xmath59 such that @xmath60 , if and only if the function @xmath61 takes its values in @xmath62 .
if @xmath6 is a retraction and @xmath63 is a classical coordinate system , then @xmath64 is the even part of a coordinate system .
conversely , if @xmath65 is a coordinate system , then there is a unique retraction @xmath6 such that @xmath66 , by the above proposition .
we call this the _ retraction associated with @xmath67 _ ( or @xmath68 )
. let @xmath69 be a coordinate system and @xmath6 be the retraction associated with @xmath67 .
any superfunction @xmath70 possesses a unique decomposition @xmath71 where @xmath72 .
observe that in the literature , one commonly writes this expansion in terms of functions @xmath73 , where @xmath74 are functions on the range of the chart associated with @xmath75 .
we note further that @xmath76 , which explains the abbreviation @xmath50 for @xmath51 .
using decomposition we define derivations along the coordinates , @xmath77 we abbreviate @xmath78 for @xmath79 and @xmath80 for @xmath81 .
corresponding abbreviations for @xmath67 and @xmath82 are similarly defined .
let @xmath3 be a supermanifold and @xmath83 a local coordinate system .
a _ berezin form _
@xmath2 on @xmath40 is an object of the form @xmath84 where @xmath70 is a superfunction on @xmath40 .
we make here no choice for the parity @xmath85 ; a common one is @xmath86 . if @xmath87 is another coordinate system on @xmath40 , then one requires @xmath88 here , the berezinian of the coordinate change is given by @xmath89 where @xmath90 thus , berezin densities have the local form @xmath91 and obey the transformation law @xmath92 where @xmath93 according to the relative orientation of @xmath75 and @xmath94
. the corresponding @xmath95-module sheaf is denoted by @xmath96 .
we denote its global sections by @xmath97 . in the literature ,
berezin forms are more common than densities . in keep with this convention
, we will use forms in .
however , in general , it will be more convenient to work with densities ; the extension to forms will always be straightforward .
in particular , this applies to the formulation and proof of our main result , in .
we recall the definition of the berezin integral .
[ def:3.3 ] let @xmath40 be a coordinate neighbourhood with a coordinate system @xmath69 , and @xmath98 .
we define @xmath99 whenever the right hand side exists . here , @xmath100 is the pullback of the standard lebesgue density on @xmath101 under @xmath75 , and @xmath102 is the top degree coefficient in equation , where @xmath6 is associated with @xmath67 . there is no uniform choice for the number @xmath103 in the literature .
customary are @xmath104 or @xmath105 .
the definition of the integral of a berezin form is similar .
we have the following classical theorem ( * ? ? ?
* theorem 2.4.5 ) .
[ thm:3.4 ] let @xmath40 be a coordinate neighbourhood and @xmath2 be a berezin density which is compactly supported on @xmath40
. then @xmath106 if @xmath69 and @xmath87 are coordinate systems on @xmath40 .
the same is true for berezin forms if @xmath75 and @xmath94 are equally oriented . as is well - known , the assumption of compact supports
can not be removed in the above theorem ; the following classical counterexample is referred to as rudakov s example in the literature .
[ ex:3.4 ] let @xmath107 be the superdomain with @xmath1080,1[$ ] .
let @xmath109 be a coordinate system on @xmath110 with @xmath111 .
let @xmath112 be the coordinate system given by @xmath113 and @xmath114 .
set @xmath115 .
we have @xmath116 hence @xmath117 .
this leads to @xmath118 however , allows us to make the following observation .
[ lem:3.5 ] let @xmath6 be a retraction on @xmath3 and @xmath2 a berezin density .
let @xmath69 and @xmath87 be coordinate systems on a coordinate neighbourhood @xmath40 with the same associated retraction @xmath6 .
let @xmath119 on @xmath40 .
then @xmath120 where @xmath102 and @xmath121 are the coefficients from equation , applied to @xmath70 and @xmath122 , respectively .
choose a bump function @xmath123 .
then by , @xmath124 since @xmath125 was arbitrary , this proves our claim .
again , one can get the same result for berezin forms .
thanks to this lemma , the following definition makes sense .
[ def:3.6 ]
let @xmath6 be a retraction on the supermanifold @xmath3 .
we define the map @xmath126 locally _ via _
@xmath127 in fact , @xmath128 can be defined for any surjective submersion @xmath129 @xcite .
note that if one chooses @xmath86 or @xmath130 and fixes parity according to the sign rule , the morphism @xmath7 becomes even .
one can easily check the following properties : @xmath131 for any @xmath132 and @xmath133 ( resp .
@xmath134 ) .
[ def:3.7 ] let @xmath6 be a retraction on @xmath3 and @xmath2 be a berezin density on @xmath3 .
we call @xmath2 integrable with respect to @xmath6 if @xmath135 is integrable on @xmath8 as density . in this case
, we define @xmath136 if @xmath8 is oriented , this definition can be extended to the case of berezin forms . on coordinate neighbourhoods @xmath40
this definition is compatible with the local definition , given in : @xmath137 where @xmath6 is the retraction associated with @xmath68 .
in particular , the integral on the right hand side is the same for coordinate systems whose even parts induce the same retraction .
moreover , generalises as follows . [ retrcpt ]
let @xmath138 be retractions on @xmath3 and @xmath2 be compactly supported on @xmath3 .
then @xmath139 in this case , we will write @xmath140 for the integral . let @xmath133 ( resp .
@xmath134 ) and @xmath138 be retractions .
the density ( resp .
volume form ) @xmath141 is exact .
if @xmath2 is compactly supported , then @xmath142 so @xmath141 is exact . in the general case ,
let @xmath143 be a partition of unity with compact supports . by the above , @xmath144 for some @xmath145 .
one may assume the family of supports to be locally finite , so that @xmath146 is well - defined , and one has @xmath147 .
let @xmath59 be an isomorphism of supermanifolds .
a. the _ pullback _ berezin density @xmath148 of a berezin density @xmath2 on @xmath17 is defined by writing @xmath149 on a coordinate neighbourhood @xmath83 on @xmath17 and setting @xmath150 here , we observe that @xmath151 is a coordinate system on @xmath152 . + this is well - defined , since @xmath153 the _ pullback _ of a berezin form is defined analogously .
b. the _ pullback _ @xmath154 of a retraction @xmath6 on @xmath17 is defined by @xmath155 [ cor:3.9 ] let @xmath59 be an isomorphism of supermanifolds .
let @xmath6 be a retraction on @xmath17 and let @xmath2 be a berezin density or berezin form on @xmath17 which is integrable with respect to @xmath6 .
then @xmath148 is integrable with respect to @xmath154 and @xmath156 in the case of a berezin form , @xmath157 has in addition to be orientation preserving .
we only have to check that @xmath158 because then @xmath159 it suffices to check equation locally .
so , we write @xmath160 and @xmath161 for a coordinate system @xmath69 with which @xmath6 is associated .
note that @xmath154 is the retraction associated with @xmath162 .
we decompose @xmath148 with respect to this coordinate system :
@xmath163 it follows that @xmath164
recall @xcite that the sheaf @xmath165 of _ integral forms _ of order @xmath166 is defined to be @xmath167 where @xmath168 denotes the @xmath169-th supersymmetric power of the sheaf of parity changed super derivations .
we will abbreviate @xmath170 by @xmath171 . in the following we restrict to the case @xmath172 .
the _ cartan derivative _ on @xmath173 integral forms is given by @xmath174 here , @xmath175 is the lie derivative on @xmath176 , locally given by @xmath177 this does not depend on the chosen coordinate system ( * ? ? ?
* lemma 2.4.6 ) . for conceptual reasons we made here a choice for the sign which differs from ref .
( the sign there is given by @xmath178 . ) [ rk:4.2 ] in the classical case @xmath179 integral forms and differential forms can be identified . for @xmath172
this identification is given by @xmath180 where @xmath181 is the contraction by @xmath182 .
the definition of the cartan derivative is compatible with this identification , as can be seen from @xmath183 recall that a morphism @xmath184 is called an _ immersion _ in case the following is true : for each point @xmath185 and some ( any ) coordinate system @xmath186 on a neighbourhood of @xmath187 , there exists a coordinate neighbourhood @xmath40 of @xmath188 , such that @xmath189 is a coordinate system on @xmath40 for certain @xmath190 .
[ la:4.3 ] if @xmath191 and @xmath192 , one can choose @xmath69 such that @xmath193 for @xmath194 and @xmath195 . for the remainder of this section we suppose @xmath17 to be of dimension @xmath196 and @xmath184 to be an immersion .
[ def:4.4 ] the _ pullback _ @xmath197 of integral forms of order @xmath173 is defined as follows : for each point @xmath198 , choose a coordinate system @xmath199 at @xmath187 as in and set @xmath200 is compatible with the classical pullback _ via _ the identification @xmath201 from .
let @xmath63 be as in ( _ i.e. _ @xmath202 ) .
one computes @xmath203 [ prop:4.6 ] the definition of the pullback at a certain point does not depend on the choice of the coordinate system and hence , the pullback of integral forms of order @xmath173 is well - defined .
let @xmath204 be another such coordinate system with @xmath205 .
we have to compute @xmath206 for @xmath207 .
for @xmath208 we infer , using @xmath209 , @xmath210 this implies @xmath211 for @xmath212 .
now we examine @xmath213 here , we have made use of @xmath214 we arrive at @xmath215 by inverting both sides of the above equation ; hence @xmath216 for the formulation of stokes s theorem , we need to anticipate a later result ( ) .
let @xmath217 such that @xmath218 has smooth boundary @xmath219 in @xmath8 .
further , let @xmath6 be a retraction on @xmath3
. then there exists a _
unique _ supermanifold structure @xmath220 of dimension @xmath196 on @xmath219 , together with an immersion @xmath221 and a unique retraction @xmath222 on @xmath220 such that the following diagram commutes : @xmath223^\iota\ar[d]_{\partial\gamma } & m\ar[d]^\gamma\\ \partial u_0\ar@{^{(}->}[r]_{\iota_0 } & m_0 } \end{xy}\end{split}\end{aligned}\ ] ] [ thm : stokes ] let @xmath217 such that @xmath224 is compact and has smooth boundary @xmath219 , and let @xmath6 be a retraction on @xmath3 .
let @xmath8 be oriented , and endow @xmath219 with the usual boundary orientation .
then for @xmath225 we have @xmath226 whenever the integral on the left hand side exists . for the special choice @xmath227
, the sign in stokes s formula disappears . therefore this choice might be reasonable in this context
we make the following subtle point : the integral on the right hand side of equation does not depend on the boundary retraction @xmath222 .
however , one still has to take into account the boundary data @xmath228 . in order to clarify this
, we consider the following example .
[ ex:4.6 ] let @xmath229 and @xmath2300,\infty[$ ] .
let @xmath231 be the standard coordinate system on @xmath3 ( @xmath232 ) .
we define another coordinate system @xmath233 by @xmath234 let @xmath6 be the retraction associated with @xmath235 , _
i.e. _ @xmath236 . in this example , there is only one possible supermanifold structure of dimension @xmath237 on @xmath238 , namely @xmath239 .
now one might think that the immersion @xmath240 is just given by @xmath241 let us examine where this leads to . define @xmath242 .
we compute @xmath243 , which implies that @xmath244 . since @xmath245 we have to calculate @xmath246 in the @xmath68-coordinates .
we see @xmath247 and @xmath248 , hence @xmath249 .
this means that @xmath250 and therefore @xmath251 the reason for this supposed contradiction is that with the chosen immersion @xmath240 , diagram does not commute .
the correct immersion is @xmath252 stokes s theorem for supermanifolds was proved @xcite for the case of domains with compact boundary .
the domain of integration there is a _ closed superdomain _ which is characterised locally by an equation @xmath253 .
this corresponds to our choice of a retraction .
the boundary of the closed superdomain is given locally by the equation @xmath254 , similar to the unique structure on the boundary , which we get from diagram . in @xcite ,
the theorem is stated as follows : one starts with a supermanifold structure on the boundary together with an immersion .
it is remarked that the boundary is given locally by an equation @xmath254 ( _ cf . _ ) , and @xmath40 by @xmath255 .
the conclusion as it is stated is correct only if the integral is evaluated by using a coordinate system which contains @xmath256 .
this means , that the integral of @xmath40 depends on the chosen immersion .
we feel that this formulation may easily be misunderstood , as in the above example , whereas the statement in terms of retractions might be more descriptive . as we shall see at the end of , stokes s theorem admits an extension to the case of an arbitrary immersion ; however , in this case , additional terms will appear in the formula . in the _ proof _ of , we need a generalisation of @xmath7 to integral forms .
define @xmath257 locally _ via _
@xmath258 here @xmath69 is a coordinate system with which @xmath6 is associated .
we check that the definition is independent of this choice . to that end , let @xmath87 be another coordinate system with @xmath236 .
then for @xmath259 we have @xmath260 , hence @xmath261 similarly , we have @xmath262 for @xmath263 .
we only need to check the equations @xmath264 with these identities , we are able to apply the classical stokes s theorem : @xmath265 the claim follows from . equations and can be checked locally .
let @xmath63 be a coordinate system such that @xmath266 and set @xmath267 .
we supplement @xmath67 to a coordinate system @xmath268 . without loss of generality
, we write @xmath269 and @xmath161 . noticing that @xmath222 is the retraction associated with @xmath270
, we get for @xmath271 : @xmath272
we will begin our examination of the behaviour of the berezin integral under coordinate changes . in view of ,
what we need to understand is how the integrals for different retractions are related .
we start with the following observation on a coordinate neighbourhood @xmath40 .
let @xmath6 and @xmath16 be retractions on @xmath40 . choose a classical coordinate system @xmath75 on @xmath218 and define @xmath267 and @xmath273 .
we complete these to coordinate systems @xmath69 and @xmath87 with @xmath274 . following , we know of the existence of a unique isomorphism @xmath275 such that @xmath276 , @xmath207 .
of course , this implies @xmath277 and @xmath278 .
if @xmath2 is a berezin density on @xmath40 , tells us that @xmath279 whenever one of both integrals exists .
one might interpret this as a first formula for coordinate changes .
however , a more explicit expression is desirable . for this reason
we take a closer look at @xmath129 .
as one can conclude from the proof of , @xmath129 is given by @xmath280 with @xmath281 and @xmath282 .
since @xmath283 , we have that @xmath284 is nilpotent for each @xmath285 , so the sum is finite . thus , @xmath286 is a differential operator of order at most @xmath287 .
there is a natural action of differential operators on berezin densities ; the following proposition can be found in ref .
@xcite .
let @xmath288 be the set of differential operators on @xmath3 .
there is a unique @xmath289-right linear action of @xmath288 on @xmath290 such that @xmath291 for all @xmath292 and @xmath293 .
the corresponding statement for @xmath176 is also correct .
note that the additional minus sign in equation can not be omitted . the so defined action is compatible with restrictions and pullbacks , _
@xmath294 where @xmath295 . in the local picture
this action has the form : @xmath296 here , @xmath297 and @xmath298 ; moreover , we set @xmath299 with this definition , the pullback _ via _ a morphism and the action of differential operators are compatible .
[ thm:5.2 ]
let @xmath300 be a isomorphism such that @xmath286 is a differential operator .
then we have for each berezin density @xmath133 @xmath301 the same is true for berezin forms , if @xmath157 is orientation preserving .
let @xmath302 be compactly supported . in our notation ,
integration by parts takes the form @xmath303 for any derivation @xmath304 . to see this , one checks
@xmath305 since @xmath306 is compactly supported , the integral of the right hand side vanishes ( * ? ? ?
* lemma 2.4.8 ) .
iteratively applying equation , we get for any @xmath307 @xmath308 using , we conclude @xmath309 since @xmath125 was arbitrary , the assertion follows .
is closely related to ref .
* theorem 3.2 ) .
it is shown there that for @xmath310 , the inverse is @xmath311 where @xmath312 for certain coordinate systems @xmath68 and @xmath235 ( where @xmath286 is denoted @xmath313 ) . in fact , equation can be deduced from equation and equation by calculating for @xmath49 @xmath314 as an aside , note that the coefficients are @xmath315 .
we use to derive an explicit expression for the berezin integral under the change of retractions .
[ thm:5.4 ] let @xmath40 be a coordinate neighbourhood with two retractions @xmath6 and @xmath16 .
let @xmath75 be a coordinate system on @xmath218 and set @xmath316 .
then @xmath317 ( or @xmath318 ) is integrable with respect to @xmath16 and @xmath319 if the right hand side exists . here
, the sum is finite , and extends over @xmath320 .
let @xmath69 , @xmath87 with @xmath321 and @xmath274 . with the morphism @xmath286 from equation ( @xmath322 ) we get @xmath323
we assemble our results in a general change of variables formula .
suppose given coordinate systems @xmath69 and @xmath87 on @xmath40 and @xmath49 .
let @xmath324 such that @xmath325 .
then @xmath326 if the integrals on the right hand side exist .
note that we eliminated the sign @xmath327 by replacing @xmath328 with @xmath329 .
further , @xmath330 always exists : @xmath331 , where @xmath16 is the retraction associated with @xmath332 .
observe that the above results bear a similarity to ref .
* theorem 3.2 ) . compared to rothstein s result , the advantage of our theorem is that it is formulated in terms of berezin densities , rather than of volume form valued differential operators . whereas the former are a locally free @xmath95-module of rank @xmath333 , the latter form one of infinite rank .
we apply these considerations in a few examples .
[ ex:5.6 ] a. recall the notation from rudakov s example ( ) . here
we have @xmath334 , hence @xmath335 . for @xmath336 and
@xmath66 our recipe shows that @xmath337 + comparing to the computations in @xmath338 , this resolves the apparent contradictions .
b. [ ex:5.6.2 ] suppose @xmath339 with @xmath340 .
let @xmath87 be a coordinate system on @xmath110 with @xmath341 .
we want to compute the @xmath235-related integral of a compactly supported @xmath342 , by using rotational symmetry .
thus , we consider on @xmath343 with @xmath344-\infty,0]\times 0)$ ] and a coordinate system @xmath69 on @xmath345 , such that @xmath346 one computes @xmath347 and @xmath348 . it remains to find @xmath330 .
+ we have @xmath349 hence @xmath350 and @xmath351 furthermore , we realise @xmath352 , hence @xmath353 .
this means that the second boundary term will vanish .
+ we write a compactly supported @xmath342 as @xmath161 on @xmath345 , where @xmath6 is associated with @xmath67 .
we obtain @xmath354 + a similar computation is contained in ( * ? ? ?
* proof of theorem 1 ) .
c. [ ex:5.6.3 ] we redo the calculation in the first example in a more general context .
suppose @xmath355 , such that @xmath356 , where @xmath3570,\infty[$ ] .
let @xmath6 and @xmath16 be retractions on @xmath345 and let @xmath358 be the standard coordinate system on @xmath359 .
then tells us for a compactly supported berezin density @xmath360 @xmath361 where @xmath267 and @xmath362 therefore , @xmath363 applying the fundamental theorem of calculus , we get @xmath364 this shows explicitly that the ` boundary terms ' indeed depend only on the values of @xmath2 , and its derivatives , on the boundary .
we shall presently exploit this to derive a global expression for the boundary terms .
in this section , we will globalise the results of the previous section , using ideas from . a framework which is well suited to such
a generalisation is that of supermanifolds built over manifolds with corners @xcite .
locally , such spaces are modelled on @xmath365 .
a family @xmath368 is called a family of _ boundary functions _ , if the @xmath14 are independent at each point at which they vanish .
this means for each subfamily @xmath369 and every @xmath370 : @xmath371 this implies that at most @xmath372 boundary functions can vanish simultaneously .
note that @xmath373 does not have to be smaller than @xmath372 ( think of the case of an interval or a rectangle ) .
observe also that @xmath369 being independent at @xmath188 implies that this family can be supplemented to a coordinate system @xmath374 on sufficiently small neighbourhoods of @xmath188 .
whenever @xmath18 is non - empty , it is called a _ boundary manifold _ of @xmath377 of codimension @xmath169 .
we set @xmath378 and denote by @xmath379 the collection of all boundary manifolds .
each boundary manifold of @xmath377 is a submanifold of @xmath8 .
furthermore , the disjoint union of all boundary manifolds coincides with the ( topological ) boundary of @xmath377 in @xmath8 .
a. the drop @xmath17 depicted in is _ not _ a manifold with corners . indeed , suppose the contrary . since there is only one codimension one boundary manifold , there can only be one boundary function .
but then @xmath17 has only one boundary manifold , and is a manifold with boundary , contradiction .
b. easy examples of manifolds with corners are displayed in .
+ a. a family of even superfunctions @xmath390 on @xmath3 is called a family of _ boundary superfunctions _ if the family of underlying functions @xmath391 is a family of boundary functions .
b. an open subspace @xmath392 is called a _
supermanifold with corners _ if there exist boundary superfunctions @xmath390 such that @xmath12 is a manifold with corners _ via _ the boundary functions @xmath393 .
let @xmath198 and @xmath394 be a subfamily of @xmath395 such that @xmath396 for each @xmath397 . similarly to the purely even case , one can augment the family @xmath394 to a coordinate system @xmath398 on a sufficiently small neighbourhood of @xmath188 .
if @xmath392 is given such that @xmath377 is a manifold with corners with boundary functions @xmath399 , we are always able to turn @xmath17 into to a supermanifold with corners _ via _ the boundary superfunctions @xmath400 for a retraction @xmath6 on @xmath3 . for the remainder of this section , @xmath17 will be a supermanifold with corners contained in the supermanifold @xmath3 ; moreover , @xmath390 will be boundary superfunctions , and @xmath375 will be boundary functions .
we call the supermanifolds @xmath19 from the _ boundary supermanifolds _ of @xmath17 corresponding to @xmath395 .
the set of all such @xmath19 is denoted by @xmath408 .
we define abbreviations @xmath409 and @xmath410 .
we will need to integrate berezin densities on @xmath3 along the boundary supermanifolds @xmath19 .
for that purpose , we define on @xmath19 canonical retractions , as well as the restriction to @xmath19 of berezin densities on @xmath3 .
[ la:5.12 ] let @xmath6 be a retraction on @xmath3 and @xmath411 .
then there exists a unique retraction @xmath21 on @xmath19 such that the following diagram commutes : @xmath412^{\iota_h}\ar[d]_{\gamma_h } & m\ar[d]^\gamma\\ h_0\ar@{^{(}->}[r]_{\iota_{h_0 } } & m_0 } \end{xy}\end{split } \end{aligned}\ ] ] for the existence , we choose @xmath415 and complete the boundary functions @xmath386 to a coordinate system @xmath416 on a sufficiently small neighbourhood @xmath417 of @xmath188 . since @xmath418 , it is clear that @xmath419 is a coordinate system on @xmath420 .
we set @xmath421 by , this defines a morphism @xmath422 such that the diagram on @xmath423 corresponding to commutes . by uniqueness
, this definition does not depend on the chosen coordinate system .
therefore , we can glue these morphisms to @xmath424 .
[ cor:5.13 ] let @xmath380 be a boundary manifold and suppose that there is another family of boundary functions @xmath425 which also determines @xmath18 .
then the families of boundary superfunctions given by @xmath426 and @xmath427 determine the same supermanifold structure @xmath19 over @xmath18 with immersion @xmath428 .
[ def:5.15 ] let @xmath431 be a boundary supermanifold .
we define a restriction map @xmath432 as follows . for @xmath415 supplement @xmath433 to a coordinate system @xmath434 on a neighbourhood @xmath40 and write @xmath435 . since @xmath436 ,
the family @xmath437 is a coordinate system on @xmath438 .
therefore , we may define @xmath439 this definition depends on @xmath395 , but it is independent of the choice of the @xmath440 ( see below ) .
thus , this defines a morphism of sheaves @xmath441 . by definition ,
@xmath442 , so it is in fact a morphism of @xmath95-modules .
a. the restriction can also be defined for berezin forms ; in this case , one has to fix an ordering on the family of boundary superfunctions @xmath395 .
+ the restriction of berezin forms and the pullback of integral forms are related as follows : suppose @xmath69 is a coordinate system on a superdomain @xmath110 .
let @xmath407 be the boundary data given by the boundary function @xmath256 .
then for any berezin form @xmath451 we have @xmath452 b. [ rk:5.16.2 ] the restriction of berezin densities is compatible with the notion of riemannian measure ; we elaborate this in the even case .
let @xmath8 carry a riemannian metric @xmath122 .
the induced riemannian density @xmath453 locally has the form @xmath454 for a coordinate system @xmath75 , where @xmath455 for @xmath456 and @xmath198 .
+ consider @xmath12 to be a manifold with corners and enumerate the boundary manifolds of codimension @xmath457 by @xmath458 .
the metric on @xmath8 determines on each @xmath459 uniquely the inner normal derivative @xmath460 . + we assume @xmath122 to be such that the @xmath459 intersect each other orthogonally .
this means that the corresponding @xmath460 are orthogonal to each other on the intersections of the @xmath459 , which are just the boundary manifolds .
thus , we can choose boundary functions @xmath375 which satisfy @xmath461 on @xmath462 .
+ let @xmath18 be any boundary manifold and @xmath415 . choose a coordinate system @xmath416 on a neighbourhood @xmath218 of @xmath188 , which satisfies @xmath463 then @xmath464 , and hence @xmath465 .
+ on the other hand , @xmath122 induces a metric @xmath466 on @xmath18 , which gives the canonical density @xmath467 on @xmath468 .
thus , @xmath469 let @xmath198 and choose a coordinate neighbourhood @xmath40 with @xmath479 .
let @xmath480 be the subfamily of @xmath395 of all boundary functions which vanish at any point of @xmath218 : @xmath481 possibly after shrinking @xmath40 , we may assume that @xmath482 can be completed to a coordinate system on @xmath40 .
since @xmath188 was arbitrary , we can choose a locally finite covering @xmath483 of @xmath3 with such coordinate neighbourhoods .
on each @xmath484 we supplement @xmath485 to a coordinate system @xmath486 and define for @xmath487 @xmath488 [ thm : trafo ] let @xmath134 be a compactly supported berezin density and let @xmath138 be retractions on @xmath3 . then @xmath509 where @xmath510 was defined in , @xmath473 is a family of boundary superderivations for @xmath426 as in , and @xmath511 note @xmath512 if @xmath513 ,
so the sum over @xmath514 is finite .
observe further that there are no summands for @xmath515 , if @xmath516 .
moreover , the berezin density @xmath517 is defined on @xmath518 , where @xmath519 .
this set contains @xmath18 , so the restriction makes sense . in general
, there is no canonical choice for the boundary superderivations ; given a super riemannian metric on @xmath3 , one might take boundary superderivations which are orthogonal to the boundary with respect to this metric ( _ cf . _ ) .
_ step 1 .
_ we suppose @xmath3 to be a superdomain , @xmath520 , and @xmath17 to satisfy @xmath521 .
the boundary functions are chosen to be @xmath389 .
furthermore , we consider a coordinate system @xmath69 with @xmath522 and @xmath66 .
a. for @xmath523 we have @xmath524 .
b. for @xmath411 and @xmath523 we have @xmath525 . c. if @xmath320 such that @xmath526 for some @xmath523 , then @xmath527 .
d. if @xmath411 and @xmath528 we get ( with @xmath529 ) @xmath530 _ step 3 .
_ now suppose @xmath3 to be a coordinate neighbourhood with boundary functions @xmath549 such that @xmath426 can be completed to a coordinate system @xmath550 .
suppose further @xmath541 to be such that @xmath542 everywhere on @xmath3 .
let @xmath551 be the inverse of a chart such that @xmath552 .
we set @xmath553 .
the restriction @xmath554 is also an isomorphism .
furthermore , @xmath555 , meaning that @xmath345 and @xmath110 are as in steps 1 and 2 .
we recognise that the boundary manifolds are sent to each other by @xmath157 , more explicitly @xmath556 indeed , @xmath157 restricts to diffeomorphisms of the boundary manifolds .
now fix @xmath411 .
we recall that @xmath437 is a coordinate system on @xmath19 .
moreover , @xmath560 is a coordinate system on @xmath561 , since @xmath562 . therefore @xmath563 defines an isomorphism @xmath564 making the following diagram commute : @xmath565^{\iota_h } & m\\ h_\omega\ar[r]_{\iota_{h_\omega}}\ar[u]^{\varphi_h } & \omega'\ar[u]_\varphi } \end{xy } \end{aligned}\ ] ] the definition of @xmath566 is also compatible with the pullback of retractions and with the restriction of berezin densities .
the former means just @xmath567 , which can be seen from the following diagram : @xmath568^{\iota_h}\ar[dd]|{\phantom{*}}^/-3ex/{\gamma_h } & & m \ar[dd]^\gamma \\
h_\omega \ar[rr]_/3ex/{\iota_{h_\omega}}\ar[dd]_{\varphi_h^*(\gamma_h)}\ar[ur]^{\varphi_h } & & \omega'\ar[dd]^/-3ex/{\varphi^*(\gamma)}\ar[ur]^/-1ex/\varphi \\ & h_0 \ar@{^{(}->}[rr]|{\phantom { * } } & & m_0 \\
\varphi_0^{-1}(h_0 ) \ar@{^{(}->}[rr]\ar[ur]_/1ex/{\varphi_{h_0 } } & & \omega'_0 \ar[ur]_{\varphi_0 } } \end{xy } \end{aligned}\ ] ] _ step 4 .
_ we prove the general formula .
let @xmath198 and choose a coordinate neighbourhood @xmath40 with @xmath479 .
denote by @xmath579 the subfamily of @xmath399 of all boundary functions which vanish at any point of @xmath218 .
let @xmath580 denote the corresponding family of boundary derivations . possibly shrinking @xmath40
we may assume that @xmath581 can be completed to a coordinate system on @xmath40 .
shrinking @xmath40 further we can suppose @xmath40 , together with @xmath579 and @xmath582 , to be as in step 3 .
since @xmath188 was arbitrary , we can take a covering @xmath583 of @xmath3 with such coordinate neighbourhoods .
let @xmath584 be a partition of unity subordinate to @xmath583 .
then @xmath585 is compactly supported in @xmath484 , hence by step 3 @xmath586 where @xmath587 and @xmath19 runs through @xmath588 .
this can be done locally .
write @xmath569 with @xmath600 . then
, similarly to step 2 of the proof of , one sees @xmath601 for some @xmath285 and @xmath602 , hence , using an analogous argument as in step 3 of the proof of , @xmath603 a. let @xmath605 be the superdomain with @xmath6060,1[^2 $ ] and @xmath138 be retractions on @xmath3 . as we have seen earlier , boundary functions for @xmath377 are given by @xmath607 .
adequate boundary derivations might be @xmath608 , where we choose @xmath609 .
+ we count the number of summands which can be non - zero . since @xmath610 we obtain @xmath457 summand for each dimension @xmath35 boundary manifold , and @xmath611 summands for each of dimension @xmath457 , resulting in @xmath612 summands .
b. [ ex:5.22.2 ] we reconsider in light of the above theorem .
recall the notation introduced in the example .
let @xmath6 be the retraction given by @xmath68 and @xmath16 be the retraction associated with @xmath235 .
we compute @xmath613 on @xmath110 :
@xmath614 + this shows that @xmath6 can be continued to a retraction on @xmath615 , where @xmath616 .
unfortunately , @xmath615 can not be considered as manifold with corners in the ambient space @xmath617 .
+ for this reason , let @xmath618 for @xmath619 be given by @xmath620 .
we turn @xmath621 to a supermanifold with corners in @xmath110 _ via _ the boundary function @xmath399 given by @xmath622 of course , @xmath623 on @xmath345 .
+ since @xmath624 , we do not need to find any boundary derivation ( although it is easy to see that the radial operator @xmath625 is a boundary derivation ) .
let @xmath342 be compactly supported . using @xmath626 we get by @xmath627 + for the application of @xmath7 we need to use @xmath628 with @xmath629 .
a simple calculation shows @xmath630 .
furthermore , we recognise that @xmath631 is the standard density on @xmath632 . in
we saw that this just means @xmath633 , which leads to @xmath634 + since @xmath635 we see again @xmath636 one may give a version of the change of variables formula for berezin forms by considering induced orientations on the boundary manifolds .
to do so , one has to fix an ordering of the boundary functions @xmath399 and has to keep track of the boundary orientations ; on the boundary manifolds of dimension @xmath35 , this leads to additional signs .
we do not state the resulting formula in full generality , since it is somewhat cumbersome .
however , in the case of a supermanifold with boundary ( _ i.e. _ for the case of only one boundary supermanifold ) , the theorem can be easily restated for berezin forms , as follows .
[ cor:5.19 ] let @xmath217 with smooth boundary @xmath219 and let @xmath138 be retractions on @xmath3 .
then for compactly supported @xmath134 we have @xmath637 here , @xmath399 is a boundary function for @xmath218 and @xmath478 is a boundary derivation corresponding to @xmath426 .
the sign @xmath638 is given by @xmath639 . in
, we saw that it is important to choose the right immersion @xmath640 to arrive at the usual formulation of stokes s theorem . in the case of an arbitrary boundary supermanifold @xmath641
, one can apply the corollary to show the following generalisation of stokes s theorem ( where , of course , additional boundary terms have to appear ) .
[ cor : stokes2 ] let @xmath217 be a supermanifold with boundary such that @xmath642 be compact , and let @xmath643 be an exact berezin form .
then @xmath644 where @xmath395 is chosen such that @xmath645 and @xmath646 . | we investigate the berezin integral of non - compactly supported quantities . in the framework of supermanifolds with corners ,
we give a general , explicit and coordinate - free repesentation of the boundary terms introduced by an arbitrary change of variables . as a corollary ,
a general stokes s theorem is derived here , the boundary integral contains transversal derivatives of arbitrarily high order .
keywords : berezin integral , stokes s theorem , change of variables , non - compact supermanifold , boundary term , manifold with corners .
msc ( 2010 ) : 58c50 , 58a50 ( primary ) ; 58c35 ( secondary ) . |
let @xmath6 be a finite subgroup . according to mckay correspondence , such a subgroup gives rise to a graph @xmath2 which turns out to be an affine dynkin diagram of ade type .
let @xmath7 be the corresponding affine lie algebra .
there are several approaches allowing one to construct @xmath7 from @xmath0 : 1 .
there is a geometric construction , due to nakajima and lusztig , of @xmath7 and its representations in terms of the quiver varieties associated to graph @xmath2 .
these varieties are closely related to the moduli spaces of instantons on the resolution of singularities @xmath8 .
2 . there is an algebraic construction , due to ringel @xcite , which allows one to get the universal enveloping algebra @xmath9 of the positive part of @xmath7 as the hall algebra of the category @xmath4 , where @xmath5 is an orientation of @xmath2 and @xmath10 is the corresponding quiver .
it was shown by peng and xiao @xcite that replacing @xmath4 by the quotient @xmath11 of the corresponding derived category , we can get a description of all of @xmath7 . different choices of orientation give rise to equivalent derived categories @xmath12 , with equivalences given by bernstein - gelfand - ponomarev reflection functors ; in terms of @xmath7 , these functors correspond to the action of the braid group of the corresponding weyl group .
recently , a third approach was suggested by ocneanu @xcite ( unpublished ) , in a closely related setting of a subgroup in quantum @xmath1 , for @xmath13 being a root of unity .
his approach is based on studying essential paths in @xmath14 , where @xmath5 is the order of the root of unity .
this approach is purely combinatorial : all constructions are done using this finite graph and vector spaces of essential paths between points in this graph , without involving any categories at all .
this paper grew out of the author s attempt to understand ocneanu s construction and in particular , find the appropriate categorical interpretation of his combinatorial constructions ; however , for simplicity we do it for subgroups in classical @xmath1 , leaving the analysis of the subgroups in quantum @xmath1 for future papers .
we show that ocneanu s essential paths in @xmath15 have a natural interpretation in terms of the category @xmath16 of @xmath0equivariant @xmath17-modules on @xmath3 ( or , rather , its `` even part '' @xmath18 , for certain natural @xmath19 grading on @xmath16 ) .
this also provides a relation with ringel lusztig construction : for ( almost ) any choice of orientation @xmath5 of @xmath2 , we construct an equivalence of triangulated categories @xmath20 these equivalences agree with the equivalences of @xmath21 for different choices of @xmath5 given by reflection functors . as a corollary
, we see that the grothendieck group @xmath22 is an affine root lattice , and the set @xmath23 of classes of indecomposable modules is an affine root system .
this construction of the affine root system via equivariant sheaves has a number of remarkable properties , namely : 1 .
this does not require a choice of orientation of @xmath2 ( unlike the category @xmath21 , where we first choose an orientation and then prove that the resulting derived category is independent of orientation ) .
2 . the indecomposable objects in the category @xmath16 can be explicitly described .
namely , they are the sheaves @xmath24 , where @xmath25 are irreducible representations of @xmath0 ( these sheaves and their translations correspond to real roots of @xmath7 ) and torsion sheaves , whose support is a @xmath0orbit in @xmath3 ( in particular , torsion sheaves whose support is an orbit of a generic point correspond to imaginary roots of @xmath7 ) .
3 . this construction of the affine root system does not give a natural polarization into negative and positive roots .
instead , it gives a canonical coxeter element in the corresponding affine weyl group , which is given by @xmath26=[{\mathcal{f}}{\otimes}{\mathcal{o}}(-2)]$ ] ( this also corresponds to the auslander
reiten functor @xmath27 ) .
4 . this construction gives a bijection of the vertices of the affine dynkin diagram @xmath2 and ( some of ) the @xmath28-orbits in @xmath23 ( as opposed to the quiver construction , where vertices of @xmath2 are in bijection with the simple positive roots ) .
for @xmath29 ( which is essentially equivalent to @xmath30 ) , these results were first obtained in the paper @xcite , where it was shown that the corresponding hall algebra contains the subalgebra isomorphic to @xmath31 . this paper in turn was inspired by an earlier paper of kapranov @xcite .
it should be noted that many of the results obtained here have already been proved in other ways .
most importantly , it had been proved by lenzing @xcite that the derived category of equivariant sheaves on @xmath3 is equivalent to the derived category of representation of the corresponding quiver ; this result has been used by schiffmann @xcite for construction of a subalgebra in the corresponding affine lie algebra via hall algebra of the category of equivariant sheaves . however , the construction of equivalence in @xcite is different than the one suggested here .
the primary difference is that in lenzing s construction , the dynkin diagram @xmath2 is constructed as the star diagram , with lengths of branches determined by the branching points of the cover @xmath32 , and he uses a standard orientation of this diagram . in the construction presented here , the diagram is @xmath2 is defined in a more standard way , using the set @xmath33 of irreducible representations of @xmath0 ; more importantly , we construct not a single equivalence but a collection of equivalences , one for each admissible orientation . in lenzing s
construction , torsion sheaves naturally play a major role ; in our construction , we concentrate on the study of locally free sheaves . lenzing
s results apply not only to @xmath34 but to a much larger class of `` non - commutative projective curves '' . however , the downside of this is that the language he uses is rather technical , making his papers somewhat hard for non - experts . for this reason ,
we had chosen to give independent proofs of some of the results , thus saving the reader the necessity of learning the language of non - commutative curves .
of course , we tried to clearly mark the results which had already been known .
* acknowledgments . *
the author would like to thank p. etingof , v. ostrik , and o. schiffmann for fruitful discussions .
special thanks to a. ocneanu , whose talk inspired this paper .
throughout the paper , we work over the base field @xmath35 of complex numbers .
@xmath0 is a finite subgroup in @xmath1 ; for simplicity , we assume that @xmath36 and denote @xmath37 ( this excludes @xmath38 , @xmath39 odd ; this case can also be included , but some of the constructions of this paper will require minor changes ) .
we denote by @xmath40 the standard 2-dimensional space , considered as a tautological representation of @xmath1 ( and thus of @xmath0 ) and @xmath41 is the @xmath42dimensional irreducible representation of @xmath1 ( and thus a representation , not necessarily irreducible , of @xmath0 ) .
we denote by @xmath43 the category of finite - dimensional representations of @xmath0 and by @xmath44 the set of isomorphism classes of simple representations ; for @xmath45 , we denote by @xmath25 the corresponding representation of @xmath0 .
since @xmath46 , the category @xmath43 is naturally @xmath19graded : @xmath47 for homogeneous object @xmath48 , we define its `` parity '' @xmath49 by @xmath50 in particular , @xmath51 .
we will also define , for @xmath45 , its parity @xmath52 .
this gives a decomposition @xmath53 we define the graph @xmath2 with the set of vertices @xmath54 and for every two vertices @xmath55 , the number of edges connecting them is defined by @xmath56 note that decomposition shows that this graph is bipartite : @xmath57 , and all edges connect vertices of different parities .
it is well - known that one can construct an isomorphism @xmath58 and that @xmath59 can be described as the space of `` essential paths '' in @xmath2 , which is naturally a direct summand in the space of all paths ( see @xcite ) .
the algebra of essential paths is also known as the preprojective algebra of @xmath2 . by mckay correspondence
, @xmath2 must be an affine dynkin diagram .
we denote by @xmath60 the corresponding affine root system ; it has a basis of simple roots @xmath61 .
we denote @xmath62 the corresponding root lattice .
it has a natural bilinear form given by @xmath63 and @xmath64 ; as is well - known , this form is positive semidefinite .
the kernel of this form is @xmath65 , where @xmath66 is the imaginary root of @xmath60 .
we denote by @xmath67 the reflection around root @xmath68 .
finally , let @xmath69 be the grothendieck group of the category @xmath43 .
it is freely generated by classes @xmath70 , i\in i$ ] ; thus , we have a natural isomorphism @xmath71&\mapsto { \alpha}_i \end{aligned}\ ] ]
we will consider a special class of orientations of @xmath2 .
[ d : height_function ] a _ height function _
@xmath5 is a map @xmath72 such that @xmath73 if @xmath55 are connected by an edge in @xmath2 , and @xmath74 .
every height function gives rise to orientation of edges of @xmath2 : if @xmath75 then all edges connecting @xmath76 and @xmath77 are directed towards @xmath77 : @xmath78 we will denote by @xmath10 the quiver given by this orientation .
we will write @xmath79 if there exists an edge whose tail is @xmath76 and head is @xmath77 .
the notation @xmath80 will mean the sum over all vertices @xmath77 connected with @xmath76 by an edge @xmath81 ; if there are multiple edges , the corresponding vertex @xmath77 will be taken more than once .
orientations obtained in this way will be called _ admissible _ ( note : our use of this word is slightly different from the use in other sources ) .
it is easy to check that if @xmath2 has no loops , then any orientation of @xmath2 is admissible . for type @xmath82 ,
an orientation is admissible if the total number of clockwise arrows is equal to the number of counterclockwise ones ( which again rules out type @xmath83 , @xmath39 even , corresponding to @xmath84 ) .
it is also obvious that @xmath10 has no oriented loops , and that adding a constant to @xmath5 gives the same orientation .
given a height function @xmath5 , we will draw @xmath2 in the plane so that @xmath5 is the @xmath85coordinate ; then all edges of @xmath10 are directed down .
figure [ f : qh ] shows an example of a height function for quiver of type @xmath86 .
[ d : reflection_functors1 ] let @xmath5 be a height function on @xmath2 , and @xmath45 be a sink in @xmath10 : there are no edges of the form @xmath79 ( in terms of @xmath5 , it is equivalent to saying that @xmath5 has a local minimum at @xmath76 ) .
we define new height function @xmath87 similarly , if @xmath76 is a source , i.e. , there are no vertices of the form @xmath81 ( equivalently , @xmath5 has a local maximum at @xmath76 ) , then we define @xmath88 one easily sees that @xmath89 , @xmath90 are again height functions ; the quiver @xmath91 is obtained from @xmath10 by reversing orientation of all edges adjacent to @xmath76
. we will refer to @xmath92 as ( elementary ) orientation reversal operations .
the following lemma is known ; however , for the benefit of the reader we included the proof .
[ l : reflection_functors1 ] any two height functions @xmath93 can be obtained one from another by a sequence of orientation reversal operations @xmath92 .
define the `` distance '' between two height functions by + @xmath94 .
we will show that if @xmath95 , then one can find @xmath92 such that @xmath92 can be applied to @xmath5 , and @xmath96 ; from this the theorem clearly follows .
let @xmath97 .
one easily sees that if @xmath98 , and @xmath81 in @xmath10 , then @xmath99 .
thus , if @xmath100 is non - empty , it must contain at least one source @xmath76 for @xmath10 .
but then @xmath101 . similarly , let @xmath102 .
if @xmath103 is non - empty , similar argument shows it must contain at least one sink @xmath76 for @xmath10 .
but then @xmath104 .
for readers convenience , we recall here the known results about the relation between representations of quiver @xmath10 and root system @xmath60 ( see , e.g. , @xcite , @xcite ) .
let @xmath5 be a height function on @xmath2 and @xmath10 the corresponding quiver .
consider the category @xmath4 of finite - dimensional representations of @xmath10 ; similarly , let @xmath105 be the bounded derived category of @xmath4 .
we denote by @xmath106 the grothendieck group of @xmath4 which can also be described as the grothendieck group of @xmath105 .
it is well - known that the category @xmath4 is a hereditary abelian category : @xmath107 for a representation @xmath108 of @xmath10 , we define its dimension @xmath109 by @xmath110 ( recall that @xmath111 $ ] is the root lattice of the root system @xmath60 , see section [ s : basic ] ) .
the following theorem summarizes some of the known results about representations of quivers and root system @xmath60 .
[ t : reps_of_quivers ] 1 .
the map @xmath112\mapsto \dim x$ ] gives an isomorphism @xmath113 . under this isomorphism ,
the bilinear form on @xmath114 is identified with the following bilinear form on @xmath106 : @xmath115 where by definition @xmath116,[y]{\rangle}= \dim \operatorname{rhom}(x , y)=\dim \operatorname{hom}(x , y ) -\dim \operatorname{ext}^1(x , y).\ ] ] 2 .
the set of dimensions of indecomposable modules is exactly the set @xmath117 of positive roots in @xmath60 . for real roots @xmath118 ,
there is exactly one up to isomorphism indecomposable module @xmath119 of dimension @xmath118 ; for imaginary root @xmath118 , there are infinitely many pairwise non - isomorphic modules of dimension @xmath118 .
there is also an explicit description of indecomposables in @xmath105 ( see ( * ? ? ?
* lemma i.5.2 ) ) .
[ t : indecompos_derived ] indecomposable objects in @xmath105 are of the form @xmath120 $ ] , where @xmath121 is an indecomposable object in @xmath4 , @xmath122
. we will also need reflection functors of bernstein gelfand ponomarev . recall that if @xmath76 is a sink in @xmath10 , then one has a natural functor @xmath123 similarly ,
if @xmath76 is a source , one has a natural functor @xmath124 ( see definition in @xcite , @xcite ) .
the following result is known and easy to prove , so we skip the proofs . [
t : coxeter2 ] 1 .
functor @xmath125 is left exact and @xmath126 is right exact . + we will denote by @xmath127 the corresponding derived functors .
2 . the functors @xmath128 , @xmath129 are equivalences of categories @xmath130 , which induce the usual reflections @xmath131 on the grothendieck group : @xmath132 3 .
if @xmath55 are not neighbors in @xmath2 , then @xmath133 commute ( i.e. , compositions in different orders are isomorphic ) and similarly for @xmath129 . in particular , for a given height function @xmath5
let @xmath134 be a sequence of elementary orientation reversal operations such that @xmath135 one easily sees that this condition is equivalent to requiring that every index @xmath45 appears in the sequence @xmath136 exactly once ; it follows from lemma [ l : reflection_functors1 ] that for every height function @xmath5 , such sequences of orientation reversal operations exist .
for such a sequence , the corresponding element of the weyl group @xmath137 is called the _ coxeter element _ , and the corresponding composition of reflection functors @xmath138 will be called the _ coxeter functor_. note that since @xmath139 as a quiver , we can consider@xmath140 as an autoequivalence of @xmath105 .
it is easy to show ( see @xcite , @xcite ) that the coxeter element @xmath141 only depends on @xmath5 and not on the choice of the sequence @xmath142 ; moreover , the proof of this only uses the fact that @xmath143 commute if @xmath55 are not connected in @xmath2 and does not use the braid relations . thus , by theorem [ t : coxeter2 ] , this implies that up to an isomorphism , @xmath144 also depends only on @xmath5 ; this justifies the notation @xmath140 .
similarly , we can define functors @xmath145 using sequences of orientation reversals @xmath146 ; the corresponding element of the weyl group will be denoted by @xmath147 .
as before , it can be shown that @xmath148 , @xmath147 only depend on @xmath5 . for readers familiar with the theory of auslander
reiten sequences ( see @xcite , @xcite ) , we add that the category @xmath4 has auslander reiten sequences , and the auslander
reiten functor @xmath27 is given by @xmath149 .
in this section , we introduce the main object of this paper , the category of equivariant sheaves on @xmath3 .
most of the results of this section are well - known and given here only for the convenience of references .
lemma [ l : frobenius ] does not seem to be easily available in the literature , but is very easy to prove . let @xmath150 be the dual of the tautological representation @xmath40 of @xmath1 . since @xmath0 is a finite subgroup in @xmath1 , it naturally acts on @xmath151 , and the structure sheaf @xmath17 has a standard @xmath1- ( and thus @xmath0- ) equivariant structure .
moreover , all twisted sheaves @xmath152 also have a standard equivariant structure , so that the space of global sections @xmath153 is a representation of @xmath0 : @xmath154 similarly , the higher homology spaces are naturally representations of @xmath0 : it is well - known that @xmath155 for @xmath156 , and @xmath157 let @xmath158 , @xmath16 be the categories of @xmath0equivariant quasi - coherent ( respectively , coherent ) @xmath17modules on @xmath3 ( see , e.g. , ( * ? ? ?
* section 4 ) for definitions ) . note that we are considering isomorphisms @xmath159 as part of the structure of the @xmath0equivariant sheaf . for brevity
, we will denote morphisms and @xmath160 groups in @xmath158 by @xmath161 , @xmath162 .
for an equivariant sheaf @xmath163 we will denote @xmath164 with the obvious equivariant structure .
similarly , for a finite - dimensional representation @xmath48 of @xmath0 , we denote @xmath165 we list here some of the basic properties of equivariant sheaves ; proofs can be found in ( * ? ? ?
* section 4 ) .
[ t : gcoh ] 1 .
@xmath158 is an abelian category , and a sequence @xmath166 is exact in @xmath158 iff it is exact in @xmath167 .
2 . for any @xmath168 , the space @xmath169 is a representation of @xmath0 , and @xmath170 . similarly , + @xmath171 ; in particular , @xmath172 for @xmath156 .
3 . for any @xmath173
, the spaces @xmath174 , @xmath175 are finite - dimensional .
4 . for any @xmath173
, one has @xmath176 as an immediate corollary of this , we see that the category @xmath16 has the krull
schmidt property : every object @xmath177 can be written as a direct sum of indecomposable modules , and the multiplicities do not depend on the choice of such decomposition .
it is also a hereditary category ( recall that a category is called hereditary if @xmath178 for any objects @xmath179 ; in particular , this implies that a quotient of an injective object is injective , as can be easily seen from the long exact sequence of @xmath160 groups ) .
we say that @xmath180 is locally free if it is locally free as an @xmath17-module .
[ t : locally_free ] 1 .
every locally free @xmath180 is isomorphic to direct sum of sheaves of the form @xmath181 2 .
if @xmath180 is locally free , then the functor @xmath182 is exact .
3 . every coherent @xmath0-equivariant sheaf has a resolution consisting of locally free sheaves .
serre duality for any two locally free sheaves @xmath183 we have isomorphisms @xmath184 it immediately follows from computation of homology of @xmath152 as coherent sheaves that @xmath185 @xmath186 finally , we note that the action of @xmath187 gives a decomposition of @xmath16 into even and odd part .
namely , since @xmath188 acts trivially on @xmath3 , structure of @xmath0-equivariant sheaf gives an isomorphism @xmath189 .
we define @xmath190 from now on , the main object of our study will be the category @xmath191 in particular , a locally free sheaf @xmath192 iff @xmath193 , where @xmath194 is the parity defined by .
one easily sees that @xmath195 is a full subcategory in @xmath16 closed under extensions .
equivalently , it can be described as follows .
let @xmath37.then @xmath195 is naturally equivalent to the category @xmath196 of @xmath197-equivariant coherent sheaves on @xmath3 .
we will denote the bounded derived category of @xmath195 by @xmath198 . for future use
, we will also need an analogue of induction functor .
recall that for any finite subgroup @xmath199 , we have an induction functor @xmath200 .
in particular , for @xmath201 and the trivial representation @xmath35 it gives the regular representation of @xmath0 : @xmath202 as any representation of @xmath0 , @xmath12 can be decomposed into even and odd part ( cf . ):
@xmath203 , where @xmath204 it is easy to check that @xmath205 is exactly the regular representation of @xmath197 .
similarly , we have an induction functor from the category of @xmath206-equivariant sheaves to @xmath0equivariant sheaves ; in particular , for @xmath201 , we get a functor @xmath207 . however , it will be more convenient to consider the functor @xmath208 the following lemma lists some of the properties of this functor ; the proof is straightforward and left to the reader .
[ l : frobenius ] let @xmath209 be defined by . then for even @xmath39 , @xmath210 is naturally isomorphic as an equivariant sheaf to @xmath211 ; for odd @xmath39 , @xmath210 is naturally isomorphic as an equivariant sheaf to @xmath212 .
the following result will play a key role in the study of the category @xmath16 . [
t : auslander_reiten ] for any @xmath122 , @xmath45 , there is short exact sequence in @xmath16 : @xmath213 where the sum is over all neighbors @xmath77 of @xmath76 in @xmath2 . we will call sequences of the form .
notice first that we have a short exact sequence of @xmath17-modules : @xmath214 let us tensor it with @xmath215 .
since tensoring with a locally free sheaf is an exact functor ( theorem [ t : locally_free ] ) , this gives short exact sequence @xmath216 since @xmath217 , we get the statement of the theorem . for readers familiar with the general theory of auslander reiten ( ar ) sequences ( see @xcite for the overview of the theory ) ,
we point out that our use of this name is justified : it can be shown , using serre duality that these sequences do satisfy the usual definition of ar sequences ( see @xcite ) . this , in particular , shows that the auslander reiten functor @xmath27 for locally free equivariant sheaves is given by @xmath218 however , we are not going to use the general theory of ar sequences in this paper . for any @xmath219 we have the following relations in the grothendieck group @xmath220 @xmath221-\sum_{j}[x_j(n+1)]+[x_i(n+2)]=0\ ] ] where the sum is over all neighbors @xmath77 of @xmath76 in @xmath2 .
later we will study the grothendieck group @xmath220 in more detail , in particular showing that relations is the full set of relations among classes @xmath222 $ ] ( see corollary [ c : generators_of_k(c ) ] ) .
in this section , we describe the indecomposable sheaves in @xmath16 . again , these results are not new ; they follow from results of @xcite , @xcite who considers more general setting of `` non - commutative projective curves '' .
so this section is included just for the reader s convenience .
we start by describing torsion sheaves . for every point @xmath223 , @xmath224 , denote @xmath225}={\mathcal{o}}/m_x^n$ ] , where @xmath226 is the ideal of functions vanishing at @xmath227 .
this sheaf is supported at point @xmath227 ; choosing a local coordinate @xmath228 at @xmath227 , we can identify the stalk of this sheaf at @xmath227 with @xmath229/z^n{\mathbb{c}}[z]$ ] .
it is well known that sheaves @xmath225}$ ] are indecomposable , and that every coherent @xmath17-module is isomorphic to a direct sum of a locally free sheaf and sheaves of the form @xmath225}$ ] .
assume now that @xmath223 is such that @xmath230 and thus @xmath231 ( recall that @xmath37 ) .
such points will be called _
generic_. define @xmath232}=\operatorname{ind}{\mathcal{o}}_{n[x]}=\bigoplus_{g\in { { \bar g}}}{\mathcal{o}}_{n[gx]}\ ] ] where the functor @xmath233 is defined by .
this sheaf is supported on the orbit @xmath234 and has a canonical @xmath197equivariant structure which can be lifted to a @xmath0-equivariant structure .
it is easy to see that the space of global sections @xmath235})$ ] is isomorphic to @xmath39 copies of the regular representation of @xmath197 .
this construction can be generalized to non - generic points .
[ t : torsion ] let @xmath223 be non - generic : @xmath236
. then 1 .
@xmath206 is a cyclic group .
2 . for suitable choice of the generator @xmath237 of @xmath206 , its action in the one - dimensional vector space @xmath238}={\mathcal{o}}/m_x$ ] is given by @xmath239 .
3 . consider the completed local ring @xmath240 ; choosing a local coordinate @xmath228 at @xmath227 identifies this algebra with @xmath241 $ ] .
it has a natural action of @xmath206 .
then we have a natural equivalence of categories @xmath242 given by @xmath243 the proof of this theorem is straightforward and left to the reader .
let @xmath244 , and @xmath245 a finite - dimensional representation of @xmath206 .
then we denote by @xmath246,y}$ ] the @xmath197-equivariant sheaf supported on the orbit @xmath234 and whose stalk at @xmath227 is isomorphic as an @xmath206-module to @xmath247 .
it is easy to show that the space of global sections of @xmath246,y}$ ] is isomorphic to @xmath248 [ t : indecomposable ] 1 .
the following is the full list of indecomposable objects in @xmath195 : * torsion sheaves @xmath246}$ ] , @xmath223 , @xmath249 , @xmath224 * torsion sheaves @xmath246,y}$ ] , @xmath223 , @xmath250 , @xmath245 an irreducible representation of @xmath206 .
* free sheaves @xmath215 , @xmath122 , @xmath45 , @xmath251 .
indecomposable objects in @xmath198 are of the form @xmath120 $ ] , @xmath121an indecomposable object in @xmath195 , @xmath122 .
there are no injective and no projective indecomposable objects in @xmath195 .
of course , sheaves @xmath246}$ ] , @xmath252 , can be considered as special case of sheaves @xmath246,y}$ ] .
however , we choose to list cases @xmath201 and @xmath253 separately . it is known that every coherent sheaf @xmath163 has a maximal torsion subsheaf @xmath254 so that @xmath255 is locally free , and short exact sequence @xmath256 splits .
if we additionally assume that @xmath163 is a @xmath197-equivariant sheaf , then @xmath254 ( and thus @xmath255 ) must also be @xmath197-equivariant .
it easily follows from theorem [ t : gcoh ] that then @xmath257 splits in the category of @xmath0-equivariant sheaves ; thus , every coherent @xmath0-equivariant sheaf is a direct sum of a free sheaf and a torsion sheaf . by theorem [ t : locally_free ] ,
indecomposable free sheaves are of the form @xmath215 . for torsion sheaves ,
it is easy to see that an indecomposable torison sheaf must be supported on an orbit @xmath234 . now
the classification of indecomposable torsion sheaves follows from theorem [ t : torsion ] .
classification of indecomposable objects in @xmath258 follows from ( * ? ? ?
* lemma i.5.2 ) .
his argument was given for the category of representations of a hereditary algebra , but actually works in any hereditary category : it only uses that a quotient of an injective object is injective , which easily follows from long exact sequence of @xmath160 spaces and the fact that @xmath259 for any @xmath260 .
we can now define a combinatorial structure which will play a crucial role in our paper .
let @xmath15 be the set of isomorphism classes of locally free indecomposable objects in @xmath195 ; by theorem [ t : indecomposable ] , we can write @xmath261 we will frequently use the notation @xmath262 we turn @xmath15 into a quiver by defining , for @xmath263 , @xmath264 , @xmath265 ( recall that @xmath266 is the number of edges between @xmath76 and @xmath77 in @xmath2 and @xmath267 is the tautological representation of @xmath1 ) .
note that the edges only connect vertices with @xmath39 differing by one , and all edges are directed `` up '' ( i.e. , so that @xmath39 increases ) .
the figure below shows an example of the quiver @xmath15 when @xmath0 is of type @xmath268 .
again , for users familiar with the auslander reiten theory , we note that @xmath15 is exactly the stable ar quiver of the category @xmath16 . as was mentioned in the introduction , it is the picture of this quiver in ocneanu s work ( who defined it purely combinatorially , without reference to the category of sheaves ) that inspired the current paper .
the following easy lemma is left as an exercise to the reader .
[ l : qhat_connected ] @xmath15 is a connected quiver .
this is actually the reason for considering the even part @xmath18 rather than all of @xmath16 : otherwise , the resulting quiver would not be connected .
note that we have a natural pairing @xmath269 given by composition @xmath270 where @xmath271 is a @xmath0-equivariant pairing ( which is defined uniquely up to a constant ) .
this gives rise to duality @xmath272 as usual in quiver theory , we define path algebra of @xmath15 by @xmath273 one easily sees that @xmath274 coincides with the space of paths in @xmath2 of length @xmath275 between @xmath77 and @xmath77 ( if @xmath276 ; otherwise , @xmath277 ) .
we also define an analogue of preprojective algebra by @xmath278 where for @xmath263 , we denote @xmath279 , the sum is over all edges originating at @xmath13 , and @xmath280 are dual bases in @xmath281 and @xmath282 with respect to the pairing .
( this algebra is isomorphic to ocneanu s algebra of essential paths , or , in the terminology of ar quivers , to the mesh algebra of @xmath15 considered as a polarized translation quiver . )
[ t : homs = paths ] let @xmath283 .
then @xmath284 let @xmath285 be the preprojective algebra of the graph @xmath2 , and @xmath286 subspace of paths of length @xmath287 .
then for @xmath288 , @xmath289 it follows from the definition that for @xmath290 we have @xmath291 and for @xmath292 , @xmath293 this theorem shows that morphisms between free indecomposable modules in @xmath195 are described by essential paths in @xmath15 , so the structure of the subcategory @xmath294 , consisting of locally free sheaves , can be recovered from @xmath15 .
note however that realization of @xmath294 in terms of equivariant sheaves gives much more than @xmath295 spaces : it embeds @xmath294 into an abelian category with sufficiently many injectives , allowing one to define @xmath160 functors and derived category @xmath198 .
both of these would be difficult to define in terms of @xmath15 .
let @xmath2 , @xmath15 be as defined above , and let @xmath5 be a height function on @xmath2 ( see definition [ d : height_function ] ) .
recall that such a height function defines a choice of orientation on @xmath2 ; the corresponding quiver is denoted @xmath10 .
then this height function defines an embedding @xmath296 where @xmath297 is the quiver obtained from @xmath10 by reversing all arrows .
the figure below shows an example of such an embedding .
note that once we have embedded @xmath298 , we can identify the quiver @xmath15 with @xmath299 ( see , e.g. , @xcite for the definition ) . from now on
, we will consider @xmath297 as a subset of @xmath15 , omitting @xmath300 in our notation .
let @xmath301 .
we say that @xmath302 is a predecessor of @xmath303 ( notation : @xmath304 ) if there exists a path @xmath305 in @xmath15 . in this case
, we also say that @xmath303 is a successor of @xmath302 and write @xmath306 .
[ l : technical1 ] let @xmath307
. then 1 .
@xmath308 for some @xmath309 . in this case
, we will say that @xmath13 is above @xmath10 and write @xmath310 .
2 . @xmath311 for some @xmath309 . in this case
, we will say that @xmath13 is below @xmath10 and write @xmath312 .
[ l : vanishing_of_homs ] 1 . if @xmath313 , then @xmath314 2 . if @xmath315 , then @xmath316
in particular , if @xmath317 , then @xmath316 .
it follows from @xmath318 that @xmath319 can be nonzero only if @xmath320 .
but then @xmath321 , which proves part ( a ) .
part ( b ) follows from part ( a ) and serre duality ( theorem [ t : locally_free ] ) [ l : generators ] the sheaves @xmath322 , @xmath323 , generate the category @xmath198 as a triangulated category : the smallest full triangulated subcategory in @xmath198 containing all @xmath322 , @xmath323 , is @xmath198 .
let @xmath324 be the smallest triangulated subcategory in @xmath198 containing all @xmath322 , @xmath323 .
if @xmath76 is a sink for @xmath10 , then it follows from ar exact sequence that @xmath325 .
thus , all @xmath322 , @xmath326 , are in @xmath324 .
similarly , if @xmath76 is a source for @xmath10 , then all @xmath322 , @xmath327 , are in @xmath324 . by lemma [ l : reflection_functors1 ] , this implies that all @xmath322 , @xmath328 , are in @xmath324 , so @xmath324 contains all locally free equivariant sheaves . since every coherent sheaf admits a locally free resolution , this completes the proof .
define now a functor @xmath329 by @xmath330 and for an edge @xmath331 in @xmath10 , we define the corresponding map by @xmath332 where @xmath333 is the edge in @xmath15 corresponding to @xmath334 under embedding .
[ l : ph ] 1 .
@xmath335 is left exact .
2 . @xmath336 for @xmath156 , and @xmath337 follows from left exactness of functor @xmath338 and definitions .
[ x : standard ] let @xmath5 be the `` standard '' height function : @xmath339 for @xmath340 , @xmath341 for @xmath342 .
then the map @xmath335 can also be described as follows .
consider the functor @xmath343 defined by @xmath344 .
then the composition @xmath345 is given by @xmath346 indeed , for each representation @xmath121 of @xmath0 we have @xmath347 applying it to @xmath348 we get @xmath349 since @xmath350 , we have @xmath351 for @xmath342 ; thus , @xmath352 similar argument shows that @xmath353 thus , @xmath354 as desired . since @xmath335 is left exact , we can define the associated derived functor @xmath355
the following two theorems are the main results of this paper .
[ t : main2 ] for any height function @xmath5 , the functor @xmath356 defined by is an equivalence of triangulated categories .
it follows from lemma [ l : vanishing_of_homs ] that the object @xmath357 satisfies @xmath358 . by lemma [ l : generators ] ,
direct summands of @xmath359 generate @xmath198 as a triangulated category .
therefore , @xmath359 is the tilting object in @xmath198 as defined in @xcite .
now the statement of the theorem follows from the general result of @xcite .
[ t : main1 ] let @xmath76 be a sink for @xmath5 , and @xmath360 the reflection functor defined by
. then the following diagram is commutative : @xmath361^{r{\phi}_{s_i^+h}}\ar[dr]^{r{\phi}_{h}}&\\ & { \mathcal{d}}^b(q_{h})\ar[uu]_{rs_i^+ } } \ ] ] and similarly for @xmath126 .
let us first prove that if @xmath310 , then @xmath362 indeed , in this case it follows from lemma [ l : ph ] , lemma [ l : vanishing_of_homs ] that @xmath363 for @xmath364 , so @xmath365 ; similarly , @xmath366 .
let @xmath76 be a sink for @xmath10 .
then we have the auslander
reiten exact sequence @xmath367 applying to this sequence functor @xmath368 and using lemma [ l : vanishing_of_homs ] which gives vanishing of @xmath369 , we get a short exact sequence @xmath370 comparing this with definition of @xmath360 , we see that @xmath371 . thus , we have shown that for @xmath310 , we have @xmath372 . to complete the proof , we now use the following easy result . [
l : technical4 ] let @xmath324 , @xmath373 be triangulated categories , @xmath374 triangle functors , and @xmath375 a morphism of functors .
assume that there is a collection of objects @xmath376 which generate @xmath324 as a triangulated category in the sense of lemma [ l : generators ] and such that for each @xmath322 , @xmath377 is an isomorphism .
then @xmath118 is an isomorphism of functors . applying this lemma to @xmath378 , @xmath379 , @xmath380 , @xmath381 and the collection of objects @xmath382 ( which generate @xmath198 by lemma [ l : generators ] )
, we get the statement of the theorem .
[ c : coxeter ] we have the following commutative diagram @xmath383 ^ -{r{\phi}_h } & { \mathcal{d}}^b(q_h)\simeq { \mathcal{d}}^b(q_{h+2})\\ { \mathcal{d}}^b({\mathcal{c}})\ar[r]^{r{\phi}_h}\ar[u]_{{\otimes}{\mathcal{o}}(-2 ) } & { \mathcal{d}}^b(q_h)\ar[u]_{rc_h^+ } } \ ] ] where @xmath140 is the coxeter element , and similarly for @xmath148 .
it should be noted that while @xmath384 is an equivalence of derived categories , it is definitely not true that @xmath335 is an equivalence of abelian categories .
for example , there are no injective or projective objects in @xmath195 while there are enough injectives and projectives in @xmath4 .
similarly , the set of simple modules is very much different in these two categories .
however , the sets of indecomposable modules are effectively the same .
in this section , we list some important corollaries of the equivalence of categories constructed in the previous section . again , many of these results can also be obtained form the equivalence of categories constructed by lenzing ; however , we choose to provide an independent exposition .
thoroughout this section , we let @xmath22 be the grothendieck group of category @xmath195 . we define the inner product on @xmath385 by @xmath386 where by definition @xmath116,[y]{\rangle}=\dim \operatorname{rhom}(x , y)=\dim \operatorname{hom}(x , y ) -\dim \operatorname{ext}^1(x , y)\ ] ] ( compare with theorem [ t : reps_of_quivers ] ) .
we define @xmath387 by @xmath388 , x \text{--- a non - zero indecomposable object in } { \mathcal{d}}^b({\mathcal{c}})\}\ ] ] finally , we define the map @xmath389 by @xmath390)=[x(-2)]\ ] ] [ t : main3 ] 1 .
the set @xmath387 is an affine root system , and @xmath28 is a coxeter element .
2 . recall the lattice @xmath114 and root system @xmath60 from section [ s : basic ] . then for any height function @xmath5 on @xmath2 , the map @xmath391&\mapsto
\bigoplus \biggl(\dim \operatorname{rhom}_{\mathcal{c}}(x_i(h_i ) , { \mathcal{f}})\biggr ) { \alpha}_i \end{aligned}\ ] ] is an isomorphism of abelian groups which identifies @xmath387 with @xmath392 and @xmath28 with the coxeter element @xmath393 defined by .
the first part follows from the second one ; the second part follows from the fact that @xmath384 is an equivalence of categories ( theorem [ t : main2 ] and corollary [ c : coxeter ] ) .
[ c : generators_of_k(c ) ] 1 . for any height
function @xmath5 , the classes @xmath394 $ ] , @xmath323 , are free generators of the grothendieck group @xmath395 .
@xmath395 is generated by the classes @xmath222 , n\equiv p(i)\mod 2 $ ] , and ar relations is the full list of all relations among classes @xmath222 $ ] . by lemma [ l : generators ]
, classes @xmath394 $ ] , @xmath323 , generate @xmath395 . on the other hand , by theorem [ t : main3 ] , @xmath395 is isomorphic to @xmath396 and thus has rank @xmath397 , so these generators must be linearly independent . to prove the second part , denote temporarily @xmath398)/j$ ] , where @xmath399 is the subgroup generated by ar relations . since @xmath222 $ ] generate @xmath395 and ar relations hold in @xmath395 , we see that @xmath395 is a quotient of @xmath400 .
now choose some height function @xmath5 .
it follows from lemma [ l : reflection_functors1 ] that @xmath394 $ ] , @xmath323 , generate @xmath400 , so it has rank at most @xmath397 . on the other hand , @xmath395 has rank @xmath397 .
thus , @xmath400 has rank @xmath397 and @xmath401 .
[ x : standard2 ] let @xmath5 be the `` standard '' height function as defined in example [ x : standard ] .
then the map @xmath402 is given by @xmath403\mapsto { \alpha}_i , \quad i \in i_0\\ & [ x_i(1)]\mapsto { \alpha}_i+\sum_{j - i}{\alpha}_j , \quad i \in i_1 \end{aligned}\ ] ] and the corresponding coxeter element is @xmath404 .
this also implies that for this @xmath5 , @xmath405\mapsto -{\alpha}_i$ ] , @xmath406 ; in particular , we see that classes @xmath407 , and @xmath408 , form a set of representatives of @xmath28-orbits on on @xmath15 .
an analogous statement for finite root system has been proved in @xcite . for completeness
, we also describe here the indecomposable sheaves corresponding to imaginary roots of @xmath23 .
[ t : imaginary_root ] let @xmath223 be a generic point : @xmath231 , and let @xmath409}]\in \delta\ ] ] see . then : 1 .
@xmath66 does not depend on the choice of point @xmath227 2 .
@xmath410 , where @xmath411=[r_0]\\ { \delta}_1&=\sum_{i\in i_1}d_i[x_i(-1)]=[r_1(-1)],\end{aligned}\ ] ] where @xmath412 and @xmath413 are even and odd parts of the regular representation defined by . 3 . @xmath414 ; @xmath415 ; @xmath416 4 . for any @xmath417 ,
@xmath418 5 .
@xmath66 is a generator of the set of imaginary roots of @xmath23 : @xmath419 we start by proving ( 2 ) , by explicitly constructing a resolution of @xmath420}$ ] .
let @xmath421 be a holomorphic section of @xmath422 which has a simple zero at @xmath227 .
then we have a short exact sequence of sheaves ( not equivariant ) : @xmath423}\to 0\ ] ] tensoring it with @xmath424 , we get a sequence @xmath425}\to 0\ ] ] now let us apply to this sequence functor @xmath233 defined by . using lemma [ l : frobenius ] , we see that it gives a @xmath197-equivariant short exact sequence @xmath426}\to 0\ ] ] which gives equality @xmath410 , thus proving part ( 2 ) of the theorem .
part ( 1 ) follows from ( 2 ) . since @xmath420}{\otimes}{\mathcal{o}}(-2)\simeq { \mathcal{o}}_{[gx]}$ ]
, we get @xmath414 . to compute @xmath427 , note that the same argument as in the proof of part ( 2 ) also gives a short exact sequence @xmath428}\to 0\ ] ] thus giving @xmath429 . to prove part ( 4 ) , recall notation @xmath430,[y]{\rangle}=\dim\operatorname{hom}(x , y)-\dim\operatorname{ext}^1(x , y)$ ]
. then serre duality immediately gives @xmath431 since @xmath414 , we get @xmath432 part ( 4 ) implies that @xmath66 is an imaginary root .
moreover , since by corollary [ c : generators_of_k(c ) ] classes @xmath70 $ ] , @xmath405 $ ] are free generators of @xmath385 .
since some of @xmath433 are equal to 1 , it follows from part ( 2 ) that for any @xmath434 , @xmath435 ; thus , @xmath66 must be a generator of the set of imaginary roots .
since every indecomposable object in @xmath198 is of the form @xmath436 $ ] , where @xmath163 is an indecomposable @xmath0-equivariant sheaf ( theorem [ t : indecomposable ] ) , it follows that every root @xmath437 can be written as either @xmath438 $ ] or @xmath439 $ ] , thus giving some splitting of @xmath440 into positive and negative roots .
this polarization can be described explicitly .
recall from algebraic geometry ( see , e.g. , ( * ? ? ?
* exercises ii.6.10 , ii.6.12 ) ) that for any coherent sheaf , we can define two integer numbers , its _
rank _ and _ degree_. in particular , for a locally free sheaf @xmath441 ( where @xmath48 is a finite - dimensional vector space ) , we have @xmath442 degree and rank give well - defined linear maps @xmath443 , where @xmath444 is the grothendieck group of the category of coherent sheaves .
in particular , we can define rank and degree for a @xmath0-equivariant sheaf , ignoring the equivariant structure ; this gives linear maps @xmath445 , which we also denote by @xmath446 .
[ l : rank ] 1 . if @xmath447 is a non - zero free sheaf , then @xmath448 .
2 . if @xmath447 is a non - zero torsion sheaf , then @xmath449 , @xmath450 .
3 . for any @xmath451 , we have @xmath452 where @xmath453 , @xmath454 are defined in theorem [ t : imaginary_root ] .
the first two parts are well - known .
to check @xmath455 , it suffices to check it for @xmath456 $ ] , @xmath340 and @xmath457 , i\in i_1 $ ] ( by example [ x : standard2 ] , they generate @xmath385 ) . if @xmath340 , then @xmath458 , { \delta}_0)=\dim \operatorname{hom}_g(x_i,\sum_{j\in i_0 } d_jx_j)=d_i=\operatorname{rk}x_i\ ] ] thus proving @xmath459 .
since @xmath460 is in the kernel of @xmath461 ( theorem [ t : imaginary_root ] ) , this implies @xmath462 . if @xmath342 , then @xmath463 , { \delta}_1)=\dim \operatorname{hom}_g(x_i(-1),\sum_{j\in i_1 } d_jx_j(-1))=d_i=\operatorname{rk}x_i\ ] ] thus proving @xmath464 .
since @xmath460 is in the kernel of @xmath461 ( theorem [ t : imaginary_root ] ) , this implies @xmath462 .
note that the functional @xmath465 can not be written in terms of the form @xmath461 : indeed , @xmath466 , but @xmath467 .
let @xmath468 . 1 . @xmath438 $ ] for some indecomposable free sheaf @xmath469 iff @xmath470 2 .
@xmath439 $ ] for some indecomposable free sheaf @xmath469 iff @xmath471 3 .
@xmath438 $ ] for some indecomposable torsion sheaf @xmath469 iff @xmath472 , @xmath473 4 .
@xmath439 $ ] for some indecomposable torsion sheaf @xmath469 iff @xmath472 , @xmath474 thus , we see that we have a triangular decomposition of @xmath440 : @xmath475,\ { \mathcal{f}}\text { --- a free indecomposable sheaf}\}\\ & \quad \text{(note that $ { \delta}_+$ is exactly the set of vertices of $ { { \widehat{q}}}$)}\\ & { \delta}_-=\{{\alpha}\in { \delta}{\ ; | \;}\operatorname{rk}({\alpha})<0\}=\{-[{\mathcal{f}}],\ { \mathcal{f}}\text { --- a free indecomposable sheaf}\}\\ & { \delta}_0=\{{\alpha}\in { \delta}{\ ; | \;}\operatorname{rk}({\alpha})=0\}=\{\pm[{\mathcal{f}}],\ { \mathcal{f}}\text { --- a free torsion sheaf}\ } \end{aligned}\ ] ] the set @xmath476 has been discussed by schiffmann @xcite , who used notation @xmath477 and denoted the corresponding subalgebra in the loop algebra by @xmath478 .
note , however , that this notation is somewhat misleading : @xmath478 is not the loop algebra of a positive part of the finite dimensional algebra @xmath479 , which easily follows from the fact that there are real roots in @xmath480 ( see example in the next section ) .
[ t : c^g ] let @xmath481 .
then for any @xmath482 we have @xmath483 let @xmath421 be a section of the sheaf @xmath422 which has a single zero at generic point @xmath227 .
then we have a short exact sequence @xmath484}\to 0\ ] ] which , however , is not equivariant even under the action of @xmath485 . to fix it we consider @xmath486 which gives the following @xmath19-equivariant sequence @xmath487}\to 0\ ] ]
now let us take product of pullbacks of @xmath486 under all @xmath488 @xmath489}\to 0\ ] ] tensoring it with any locally free sheaf @xmath163 , we get @xmath490}{\otimes}{\mathcal{f}}_x\to 0\ ] ] which implies @xmath491-[{\mathcal{f}}]+2(\operatorname{rk}{\mathcal{f}}){\delta}=0 $ ] .
this result that @xmath492 is a translation was known before and can be proved without the use of equivariant sheaves , see e.g. steinberg @xcite .
however , the approach via equivariant sheaves also provides a nice interpretation for the corresponding functional as the rank of the sheaf . comparing with the description of the action of coxeter element in the language of representations of the quiver
, we see that rank is closely related to the notion of defect @xmath493 as defined in @xcite , namely @xmath494 therefore , @xmath480 is exactly the set of indecomposable objects of defect zero , which shows that torsion sheaves correspond to regular representations . 1 . for any @xmath495 , @xmath496 ;
in particular , @xmath28-orbit of @xmath118 is finite .
@xmath28 acts freely on @xmath476 , and the set of orbits @xmath497 is naturally in bijection with @xmath2 .
in this section , we consider the example of the cyclic group of even order : @xmath38 , @xmath498 . the root system @xmath60 can be described as follows .
let @xmath40 be a real vector space of dimension @xmath500 , with basis @xmath501 , @xmath502 .
define inner product in @xmath40 by @xmath503 , @xmath504
. then @xmath505 the simple roots are @xmath506 so that @xmath507 .
the simple reflections @xmath131 are given by @xmath508 we choose standard height function @xmath5 : @xmath510 the corresponding coxeter element @xmath28 is @xmath511 the action of @xmath28 on @xmath60 is given by @xmath512 thus , we have @xmath513 which implies @xmath514 ( compare with theorem [ t : c^g ] , lemma [ l : rank ] ) .
w. geigle , h. lenzing , _ a class of weighted projective curves arising in the representation theory of finite - dimensional algebras _ , in _ singularities , representations of algebras and vector bundles ( lambrecht , germany , 1985 ) _ , lecture notes in math .
1273 , springer , berlin , 1987 , 265297 h. kraft , ch .
riedtmann , _ geometry of representations of quivers _ , in _ representations of algebras ( durham , 1985 ) _ , 109145 , london math .
lecture note ser .
, 116 , cambridge univ . press , cambridge , 1986 . c. m. ringel , _ representation theory of finite - dimensional algebras _ , in _ representations of algebras ( durham , 1985 ) _ , 779 , london math . soc .
lecture note ser .
, 116 , cambridge univ .
press , cambridge , 1986 . | let @xmath0 be a finite subgroup in @xmath1 , and @xmath2 the corresponding affine dynkin diagram . in this paper
, we review the relation between the categories of @xmath0-equivariant sheaves on @xmath3 and @xmath4 , where @xmath5 is an orientation of @xmath2 , constructing an explicit equivalence of corresponding derived categories . |
spin net gravity is based on the quantization of gravity using connection variables .
( for a review , see ref .
@xcite . ) searching for the eigenspace for geometric observables , rovelli and smolin introduced new states @xcite which echoed an older , combinatorial definition of spacetime advocated by penrose @xcite . using penrose s original name
the new states were named `` spin networks '' .
these were later shown to be a basis @xcite for the kinematic or 3-geometry states of quantum gravity . but
despite this success , the dynamics remains controversial .
`` spin net gravity '' makes use of the kinematic states and includes the assumption that the spin network states are robust enough to support the full dynamics of quantum gravity .
spin networks are discrete , graph - like structures .
a spin network , @xmath1 , is defined as the triple @xmath2 of an oriented graph , @xmath3 , labels on the vertices ( or `` intertwiners '' ) , @xmath4 , and integer edge labels , @xmath5 .
the edge labels index the irreducible representation carried on the edge . the corresponding spin net state @xmath6 is defined in the connection representation as @xmath7,\ ] ] where the holonomy @xmath8 along edge @xmath9 is in the @xmath10 irreducible representation of @xmath11 .
when the intertwiners `` connect '' all the incident edges .
when the intertwiners are invariant tensors on the group , the states are gauge invariant .
one may represent intertwiners of higher valence vertices as a decomposition into trivalent vertices and labels on the `` internal '' edges which connect these trivalent vertices .
spin networks are the eigenspace of quantum operators which measure geometric quantities such as length , area , volume , and angle .
for instance , when an edge @xmath9 of a spin network passes through a surface , it contributes area @xmath12 to the area of that surface @xcite .
hence , edges are called `` flux lines of area '' . vertices support volume and angle and may be called `` atoms of geometry '' .
the new quantum geometry is completely different than our everyday , continuous geometry .
in fact , this theory predicts that space is fundamentally discrete . in the next section
an angle operator is introduced and its semi - classical limit is discussed . a quasilocal energy operator is briefly reviewed in section [ quasi ] .
the report is summarized in section [ sum ] .
the idea that one could measure angle from the discrete structure of spin networks first arose in the context of penrose s work @xcite .
this work culminated in the spin geometry theorem , which stated that angles in 3-dimensional space could be approximated to arbitrary accuracy if the spin network was sufficiently correlated and if the spins were sufficiently large .
this result is an extension of the observation that , for an epr pair , we know everything about the relative orientation of the particles spins but nothing about the absolute orientation of the two spins in space .
the spin geometry construction carries over into the context of quantum geometry : one can measure the angle between two ( internal ) edges of a spin network state .
the angle operator is a quantization of the classical expression for an angle @xmath13 measured at a point @xmath14 @xmath15 with surface normals @xmath16 and @xmath17 for surfaces @xmath18 and @xmath19 .
the regularization of the classical expression is straightforward @xcite .
the resulting operator is very similar to the operator used by penrose in the diagrammatic form of the spin geometry theorem .
it assigns an angle to two bundles of edges incident to a vertex as shown in fig .
[ cones ] .
the quantum operator is defined as follows : incident edges are partitioned into three categories which correspond to the two cones shown in fig .
( [ cones ] ) and the remainder .
all of the edges in each category connect to a single internal edge in the `` intertwiner core . ''
these edges are denoted @xmath20 , @xmath21 , and @xmath22 . for reasons that will be clear in a moment , the spin @xmath22 is known as the `` geometric support '' .
associated to these partitions and edges are three spin operators @xmath23 , @xmath24 , and @xmath25 .
thus , @xmath26^{1/2 } } \right ) { \mid s \rangle}. \end{split}\ ] ] in the second line , the intertwiner core with a trivalent vertex labeled by @xmath20 , @xmath21 and @xmath22 is used .
the key idea of the angle operator is to measure the relative spins of internal edges coming from two disjoint conical regions .
the property which sets this angle operator apart from the other geometric observables is the complete absence of the scale of the theory @xmath27 .
the result is _ purely _ combinatorial ! however , there is an important technical caveat : if one retains the diffeomorphism invariance of the classical theory in the construction of the state space , then there is , generically , a set of continuous parameters or moduli space associated with higher valence vertices @xcite .
these parameters contain information on the embedding of the spin network graph in space .
on such a space the angle spectrum is highly degenerate .
it is by no means clear whether this embedding information is physically relevant @xcite .
indeed , the state space of quantum gravity may be described by abstract or non - embedded graphs .
the spectrum of the angle operator is relatively easy to calculate for small spins @xcite . a useful parameter to characterize
this spectrum is the `` spin sum '' @xmath16 , @xmath28 .
the first few cases are given in table 1 .
two aspects of the low - lying spectrum are immediately obvious . while the examples in table 1 include 180 degrees ,
there is a large gap between the smallest angle and 0 degrees ; it is `` hard '' to model small angles .
second , the difference between neighboring angles is non - uniform .
this is especially apparent in the @xmath29 case .
the difference between 0 and the lowest angle is 50 degrees while neighboring angles are as close as @xmath30 degrees .
further , neighboring angles do not correspond to a simple numerical progression of the internal spins @xmath31 and @xmath22 .
the combinatorial nature and the non - uniform spacing of the spectrum means that the semi - classical limit of the operator offers a wealth of information about the nature of the spin network vertex .
the semi - classical limit is not merely a matter of reducing the `` grain '' of the fundamental geometry . not only must we match the current experimental bounds on the smoothness of angle but we must also find small angles and match the classical distribution of angles in 3-space .
small angles are ` hard ' to achieve in the following sense : to support an angle near zero the argument of the arc cosine in eq .
( [ aop ] ) must be close to unity .
that is , we must maximize the argument .
but the edge labels satisfy the triangle inequalities . as shown in ref .
@xcite a short argument shows that , for a small angle @xmath32 , @xmath33 since the total spin is bounded from above by twice the spin fluxes through the cones , the minimum observable angle depends on the area flux through the regions we are measuring . for instance , to achieve an angle as small as @xmath34 radians , the required flux is roughly @xmath35 .
c c c | r + @xmath22 & @xmath20 & @xmath21 & @xmath36 ( degrees ) + 2 & 1 & 1 & 70.53 + 1 & 1 & 2 & 144.74 + 0 & 2 & 2 & 180.00 + number of distinct angles : 3 c c c | r + @xmath22 & @xmath20 & @xmath21 & @xmath36 ( degrees ) + 3 & 1 & 2 & 65.91 + 2 & 2 & 2 & 120.00 + 2 & 1 & 3 & 138.19 + 1 & 2 & 3 & 155.91 + 0 & 3 & 3 & 180.00 + number of distinct angles : 5 c c c | r + @xmath22 & @xmath20 & @xmath21 & @xmath36 ( degrees ) + 4 & 2 & 2 & 60.00 + 4 & 1 & 3 & 63.43 + 3 & 2 & 3 & 111.42 + 3 & 1 & 4 & 135.00 + 2 & 3 & 3 & 137.17
+ 2 & 2 & 4 & 150.00 + 1 & 3 & 4 & 161.57 + 0 & 4 & 4 & 180.00 + number of distinct angles : 8 c c c | r + @xmath22 & @xmath20 & @xmath21 & @xmath36 ( degrees ) + 5 & 2 & 3 & 56.79 + 5 & 1 & 4 & 61.87 + 4 & 3 & 3 & 101.54 + 4 & 2 & 4 & 106.78 + 3 & 3 & 4 & 129.23 + 4 & 1 & 5 & 133.09 + 2 & 4 & 4 & 146.44 + 3 & 2 & 5 & 146.79 + 2 & 3 & 5 & 156.42 + 1 & 4 & 5 & 165.04 + 0 & 5 & 5 & 180.00 + number of distinct angles : 11 c c c | r + @xmath22 & @xmath20 & @xmath21 & @xmath36 ( degrees ) + 6 & 3 & 3 & 53.13 + 6 & 2 & 4 & 54.74 + 6 & 1 & 5 & 60.79 + 5 & 3 & 4 & 96.05 + 5 & 2 & 5 & 103.83 + 4 & 4 & 4 & 120.00 + 4 & 3 & 5 & 124.57 + 5 & 1 & 6 & 131.81 + 3 & 4 & 5 & 139.38 + 4 & 2 & 6 & 144.74 + 2 & 5 & 5 & 152.34 + 3 & 3 & 6 & 153.43 + 2 & 4 & 6 & 160.53 + 1 & 5 & 6 & 167.40 + 0 & 6 & 6 & 180.00 + number of distinct angles : 15 c c c | r + @xmath22 & @xmath20 & @xmath21 & @xmath36 ( degrees ) + 7 & 3 & 4 & 50.77 + 7 & 2 & 5 & 53.30 + 7 & 1 & 6 & 60.00 + 6 & 4 & 4 & 90.00 + 6 & 3 & 5 & 92.50 + 6 & 2 & 6 & 101.78 + 5 & 4 & 5 & 114.46 + 5 & 3 & 6 & 121.45 + 6 & 1 & 7 & 130.89 + 4 & 5 & 5 & 131.08 + 4 & 4 & 6 & 135.00 + 5 & 2 & 7 & 143.30 + 3 & 5 & 6 & 146.05 + 4 & 3 & 7 & 151.44 + 2 & 6 & 6 & 156.44 + 3 & 4 & 7 & 157.79 + 2 & 5 & 7 & 163.40 + 1 & 6 & 7 & 169.11 + 0 & 7 & 7 & 180.00 + number of distinct angles : 19 the angle spectrum must also match the behavior of angles in three - dimensional space . in the classical , smooth model of angles the distribution of solid angles is proportional to @xmath37 .
we expect that the discrete model of space should be able to reproduce this behavior in the appropriate limit .
therefore , the classical limit not only encompasses the need to achieve small angles and fine angular resolution , but also the need to match the distribution of solid angles @xcite . in the discrete model
this reduces to finding the number of intertwiners with a given core .
this distribution can be found numerically .
the results are shown in fig .
( [ dist ] ) for a choice of surface fluxes .
details of this calculation are in ref .
@xcite .
though it is not obvious by looking at this one plot , to match the expected distribution of solid angle it is necessary that the geometric support @xmath22 be many orders of magnitude larger than the surface fluxes through @xmath18 and @xmath19 .
further , the spin flux through the two surfaces must be approximately the same magnitude .
on spatial manifolds without boundary , the action for the classical ( 3 + 1 ) form of gravity is a set of constraints .
suppose that the space does have a boundary @xmath38 .
under the boundary conditions @xmath39 the theory is inconsistent without the boundary term @xmath40 ( see ref .
@xcite or , for a more general setting , ref . @xcite . )
this is the hamiltonian for the system .
the quantization of this observable is carried out in ref .
the most general form of the operator is @xmath41 the first sum is over all the intersections @xmath14 of the surface with the spin network .
the second sum is over all the incident edges @xmath42 and @xmath43 .
@xmath44 is the lapse at the intersection @xmath14 ; @xmath45 is a sign factor for the edges @xmath42 ; and @xmath46 is the area operator .
the spectrum is most easily stated for two types of intersections .
type ( i ) intersections are transversal while type ( ii ) include all other intersections . for higher valence vertices ,
the intertwiner core is labeled with edges @xmath47 and @xmath48 corresponding to incident edges @xmath42 with positive , negative and zero values of @xmath45 .
for the operator defined in eq .
( [ qeadef ] ) , one has for type ( i ) @xmath49{n_{v}(n_{v}+2 ) } { \mid s \rangle}. \\ \text{for type ( ii):}\\ \hat{h}_{\partial \sigma}(n ) { \mid s \rangle } & = m \sum_{v \in s \cap { \mathsf g } } n_{v } \frac{p_{v}(p_{v}+2 ) + n_{v}(n_{v}+2 ) - z_{v}(z_{v}+2 ) } { \left [ 2p_{v}(p_{v}+2 ) + 2n_{v}(n_{v}+2 ) - z_{v}(z_{v}+2 ) \right]^{\frac{3}{4 } } } { \mid s \rangle}. \end{split}\ ] ] the fundamental mass scale is defined as @xmath50 this is essentially the same scale as @xmath27 of the geometric observables .
there are two remarks to make : for type ( i ) intersections @xmath14 , the quasilocal energy operator has eigenvalue @xmath51 , which is related to the area as @xmath52 in which @xmath53 is the eigenvalue of the area operator .
the quasilocal energy is a sum of contributions from each intersection ; the energy is a sum of independent , non - interacting particle - like excitations .
this vastly simplifies the statistical mechanics based on this hamiltonian @xcite .
there are two new operators for spin net gravity , angle and quasilocal energy . with length , area , and volume
they provide a number of ways in which we are able to explore this discrete model of space .
the angle operator is the first geometric operator that is purely combinatorial ; it depends only on the spins of the vertex , not on the scale @xmath27 .
this also means that it is free of the quantization ambiguity associated with the immirzi parameter @xcite .
further , the quantization of this operator suggests that scalar products also have discrete spectra .
though this has largely escaped notice , this obviously has far reaching consequences in a variety of high and low energy physics . in the case of the angle
, we learn that the semi - classical limit is not merely determined by large spins , but by large spins on the internal edges of the intertwiner located at the vertex . to correctly model the distribution of angles we require not only that these spins be large ( @xmath54 ) but also that the geometric support be many orders of magnitude larger ( @xmath0 ) .
both operators share what appears to be a common feature with all operators in spin net gravity , the spectra are discrete .
what is utterly remarkable is not just this - the discrete spectra of geometric quantities - but what this implies for _ all _ of space , whether it be curved or flat .
if this model is correct , fundamental quantum geometry affects flat - space physics .
we have only just begun to realize the consequences of such a dramatic shift in the foundations of spacetime geometry .
roger penrose , `` angular momentum : an approach to combinatorial spacetime '' in _ quantum theory and beyond _ edited by t. bastin ( cambridge university press , cambridge , 1971 ) ; `` combinatorial quantum theory and quantized directions '' in _ advances in twistor theory _ , research notes in mathematics 37 , edited by l. p. hughston and r. s. ward ( pitman , san fransisco , 1979 ) pp .
301 - 307 ; in _ combinatorial mathematics and its application _ edited by d. j. a. welsh ( academic press , london , 1971 ) ; `` theory of quantized directions , '' unpublished notes . | two operators for quantum gravity , angle and quasilocal energy , are briefly reviewed .
the requirements to model semi - classical angles are discussed . to model semi - classical angles
it is shown that the internal spins of the vertex must be very large , @xmath0 . |
cosmic radio jets are believed to be originated from the centers of active galaxies which harbor black holes ( e.g. , chakrabarti 1996a , hereafter c96a ) . even in so called ` micro - quasars ' , such as grs 1915 + 105 which are believed to have stellar mass black holes ( mirabel & rodriguez , 1994 )
, the outflows are common .
the well collimated outflows in ss 433 is well known for almost two decades ( margon , 1984 ) .
similarly , systems with neutron stars also show outflows as is believed to be the case in x - ray bursters ( e.g. titarchuk , 1994 ) .
there are large number of works in the literature which attempt to explain the origin of these outflows .
these works can be broadly divided into three sets . in one set
, the jets are believed to come out due to hydrodynamic or magneto - hydrodynamic pressure effects and are treated separately from the disks ( e.g. , blandford & payne 1982 ; fukue 1982 ; chakrabarti 1986 ; contopoulos 1995 ) . in another set
, efforts are made to correlate the disk structure with that of the outflow ( e.g. , knigl 1989 ; chakrabarti & bhaskaran 1992 ) . in the third set ,
numerical simulations are carried out to actually see how matter is deflected from the equatorial plane towards the axis ( e.g. , hawley , smarr & wilson 1985 ; eggum , coroniti & katz , 1985 ; molteni , lanzafame & chakrabarti 1994 ; molteni , ryu & chakrabarti 1996 ) . nevertheless ,
the definitive understanding of the formation of outflows is still lacking , and more importantly , it has always been difficult to estimate the outflow rate from first principles . in the first set ,
the outflow is not self - consistently derived from the inflow . in the second
set , only self - similar steady solutions are found and in the third set , either a keplerian disk or a constant angular momentum disk was studied , neither being the best possible assumption . on the other hand ,
the mass outflow rate of the normal stars are calculated very accurately from the stellar luminosity .
theory of radiatively driven winds seems to be very well understood ( e.g. , castor , abott & klein , 1975 ) .
given that the black holes and the neutron stars are much simpler celestial objects , and the flow around them is sufficiently hot to be generally ionized , it should have been simpler to compute the outflow rate from an inflow rate than the methods employed in stellar physics .
our approach to the mass outflow rate computation is somewhat different from that used in the literature so far .
though we consider simple minded equations to make our points , such as those applicable to conical inflows and outflows , we add a fundamental ingredient to the system , whose importance is being revealed only very recently in the literature .
this is the quasi - spherical centrifugally supported dense atmosphere of typical size @xmath0 to @xmath1 schwarzschild radius around a black hole and a neutron star . whether a shock actually forms or not , this dense region exists , as long as the angular momentum of the flow close to the compact object is roughly constant and is generally away from a keplerian distribution as is the case in reality ( chakrabarti , 1989 [ hereafter c89 ] ; chakrabarti 1996b ;
hereafter c96b ) .
this centrifugally supported region ( which basically forms the boundary layer of black holes and weakly magnetized neutron stars ) successfully replaced the so called ` compton cloud ' ( chakrabarti & titarchuk 1995 [ hereafter ct95 ] ; chakrabarti , titarchuk , kazanas & ebisawa 1996 ; chakrabarti , 1997 [ hereafter c97 ] ) in explaining hard and soft states of black hole candidates , and the converging flow property of this region successfully produced the power - law spectral slope in the soft states of black hole candidates ( ct95 ; titarchuk , mastichiadis & kylafis 1997 ) .
the oscillation of this region successfully explains the general properties of the quasi - periodic oscillation ( molteni , sponholz & chakrabarti , 1996 ; ryu , chakrabarti & molteni 1997 ) from black holes and neutron stars .
it is therefore of curiosity if this region plays any major role in formation of outflows .
several authors have also mentioned denser regions due to different physical effects .
chang & ostriker , 1985 showed that pre - heating of the gas could produce standing shocks at a large distance .
kazanas & ellison ( 1986 ) mentioned that pressure due to pair plasma could produce standing shocks at smaller distances around a black hole as well .
our computation is insensitive to the actual mechanism by which the boundary layer is produced .
all we require is that the gas should be hot at the region where the compression takes place .
thus , since comptonization processes cool this region ( ct95 ) for larger accretion rates ( @xmath2 ) our process is valid only for low - luminosity objects , consistent with current observations .
begelman & rees ( 1984 ) talked about a so - called ` cauldron ' model of compact objects where jets were assumed to emerge from a dense mixture of matter and radiation by boring de - laval nozzle as in blandford & rees ( 1974 ) model .
the difference between this model and the present one is that there very high accretion rate was required ( @xmath3 ) while we consider thermally driven outflows of smaller accretion rate .
second , the size of the ` cauldron ' was thousands of schwarzschild radii ( where gravity was so weak that channel has to have shape of a de laval nozzle ) , while we have a cenbol of about @xmath4 ( where the gravity plays an active role in creating the transonic wind ) in our mind .
third , in the present case , matter is assumed to pass through a sonic point using the pre - determined funnel where rotating pre - jet matter is accelerated ( chakrabarti , 1984 ) and not through a ` bored nozzle ' even though symbolically a quasi - spherical cenbol is considered for mathematical convenience .
fourth , for the first time we compute the outflow rate completely analytically starting from the inflow rate alone . to our knowledge
such a calculation has not been done in the literature at all .
once the presence of our centrifugal pressure supported boundary layer ( cenbol ) is accepted , the mechanism of the formation of the outflow becomes clearer .
one basic criteria is that the outflowing winds should have positive bernoulli constant ( c89 ) . just as photons from the stellar surface deposit momentum on the outflowing wind and keeps the flow roughly isothermal ( tarafdar , 1988 ) at least upto the sonic point , one may assume that the outflowing wind close to the black hole is kept isothermal due to deposition of momentum from hard photons . in the case of the sun
, it s luminosity is only @xmath5 and the typical mass outflow rate from the solar surface is @xmath6 year@xmath7 ( priest , 1982 ) .
proportionately , for a star with a eddington luminosity , the outflow rate would be @xmath8 year@xmath7 .
this is roughly half as the eddington rate for a stellar mass star .
thus if the flow is compressed and heated at the centrifugal barrier around a black hole , it would also radiate enough to keep the flow isothermal ( at least up to the sonic point ) if the efficiency were exactly identical .
physically , both requirements may be equally difficult to meet , but in reality with photons shining on outflows near a black hole with almost @xmath9 solid angle ( from funnel wall ) it is easier to maintain the isothermality in the slowly moving ( subsonic ) region in the present context .
another reason is this : the process of momentum deposition on electrons is more efficient near a black hole .
the electron density @xmath10 falls off as @xmath11 while the photon density @xmath12 falls off as @xmath13.thus the ratio @xmath14 increases with the size of the region .
thus a compact object will have lesser number of electrons per photon and the momentum transfer is more efficient . in a simpler minded way
the physics is scale - invariant , though . in solar physics , it is customary to chose a momentum deposition term which keeps the flow isothermal to be of the form ( kopp & holzer , 1976 ] , @xmath15 where , @xmath16 is the momentum deposition ( localized around @xmath17 ) factor with a typical spatial dependence , @xmath18 here , @xmath19 , @xmath20 are constants and @xmath21 is the location of the stellar surface . since @xmath22 and @xmath17 comes in ratio , exactly same physical consideration would be applicable to black hole physics , with the same result _ provided _
@xmath19 is scaled with luminosity ( however , as we showed above , @xmath19 goes up for a compact object . ) .
however , as ct95 showed , high accretion rate ( @xmath23 ) will _ reduce _ the temperature of the cenbol catastrophically , and therefore our assumption of isothermality of the outflow would severely breakdown at these high rates .
it is to be noted that in the context of stellar physics , it is shown ( pauldrachi , puls & kudritzki , 1986 ) that the temperature stratification is the outflowing wind has little effect on the mass loss rate .
having thus convinced that isothermality of the outflow , at least upto the sonic point , is easier to maintain near a black hole , we present in this _ paper _ a simple derivation of the ratio of the mass outflow rate and mass inflow rate assuming the flow is externally collimated .
we find that the ratio is a function of the compression ratio of the gas at the boundary of the hot , dense , centrifugally supported region .
we estimate that the outflow rate should generally be less than a few percent if the outflow is well collimated .
finally , in 3 , we draw our conclusions .
assume for the sake of argument that our system is made up of the infalling gas , the dense boundary layer of the compact object ( cenbol ) , and collimated outflowing wind .
figure 1 shows a schematic diagram of the system with the components marked .
matter near the equatorial plane is assumed to fall in conical shape onto the black hole or the neutron star .
the sub - keplerian , hot and dense , quasi - spherical region forms either due to centrifugal barrier or due to pair plasma pressure or pre - heating effects .
the outflowing wind is assumed to be also conical in shape for simplicity and is flowing out along the axis .
it is assumed that the wind is collimated by an external pressure .
both the inflow and the outflow are assumed to be thin enough so that the velocity and density variations across the flow could be ignored .
the accretion rate of the incoming accretion flow is given by , @xmath24 here , @xmath25 is the solid angle subtended by the inflow , @xmath26 and @xmath27 are the density and velocity respectively , and @xmath22 is the radial distance . for simplicity , we assume geometric units ( @xmath28 ; @xmath29 is the gravitational constant , @xmath30 is the mass of the central black hole , and @xmath31 is the velocity of light ) to measure all the quantities . in this unit , for a freely falling gas , @xmath32^{1/2 } \eqno{(2)}\ ] ] and @xmath33 here , @xmath34 ( with @xmath35 assumed to be a constant ) is the outward force due to radiation .
we assume that the boundary of the denser cloud is at @xmath36 ( typically a few schwarzschild radii , see , chakrabarti 1996b ) where the inflow gas is compressed . the compression could be abrupt due to standing shock or gradual as in a shock - free flow with angular momentum .
this details are irrelevant . at this barrier
, then @xmath37 and @xmath38 where , @xmath39 is the compression ratio . exact value of the compression ratio is a function of the flow parameters , such as the specific energy and the angular momentum ( e.g. , c89 ; chakrabarti 1990 [ hereafter c90 ] , chakrabarti , 1996c ) . here ,
the subscripts @xmath40 and @xmath41 denote the pre - shock and post - shock quantities respectively . at the shock surface , the total pressure ( thermal pressure plus ram pressure )
is balanced .
@xmath42 assuming that the thermal pressure of the pre - shock incoming flow is negligible compared to the ram pressure , using eqs .
4(a - b ) we find , @xmath43 the isothermal sound speed in the post - shock region is then , @xmath44 where , @xmath45 .
an outflow which is generated from this dense region with very low flow velocity along the axis is necessarily subsonic in this region , however , at a large distance , the outflow velocity is expected to be much higher compared to the sound speed , and therefore the flow must be supersonic . in the subsonic region of the outflow ,
the pressure and density are expected to be almost constant and thus it is customary to assume isothermality condition up to the sonic point ( tarafdar , 1988 ) . as argued in the introduction , in the case of black hole accretion also , such an assumption may be justified . with isothermality assumption or a given temperature distribution ( @xmath46 with @xmath47 a constant )
the result is derivable in analytical form .
the sonic point conditions are obtained from the radial momentum equation , @xmath48 and the continuity equation @xmath49 in the usual way , i.e. , by eliminating @xmath50 , @xmath51 where @xmath52 and @xmath53 and putting @xmath54 and @xmath55 conditions .
these conditions yield , at the sonic point @xmath56 , for an isothermal flow , @xmath57 and @xmath58 where , we have utilized eq .
( 7 ) to substitute for @xmath59 .
since the sonic point of a hotter outflow is located closer to the black hole , clearly , the condition of isothermality is best maintained if the temperature is high enough .
however if the temperature is too high , so that @xmath60 , then the flow has to bore a hole through the cloud just as in the ` cauldron ' model of begelman & rees ( 1984 ) , although it is a different situation here the temperature is high , while in the ` cauldron ' model the temperature was low . in reality , a pre - defined funnel caused by centrifugal barrier does not require to bore any nozzle at all , but our simple quasi - spherical calculation fails to describe this case properly .
this is done in detail in das & chakrabarti ( submitted ) .
the constancy of the integral of the radial momentum equation ( eq . 8) in an isothermal flow gives : @xmath61 where , we have ignored the initial value of the outflowing radial velocity @xmath62 at the dense region boundary , and also used eq .
we have also put @xmath63 and @xmath64 . upon simplification , we obtain , @xmath65 where , @xmath66 thus , the outflow rate is given by , @xmath67 where , @xmath68 is the solid angle subtended by the outflowing cone .
upon substitution , one obtains , @xmath69 which , explicitly depends only on the compression ratio : @xmath70^{3/2 } exp ( \frac{3}{2 } - \frac{r^2}{r-1 } ) \eqno{(16)}\ ] ] apart from the geometric factors . notice that this simple result does not depend on the location of the sonic points or the the size of the dense cloud or the outward radiation force constant @xmath35 .
this is because the newtonian potential was used throughout and the radiation force was also assumed to be very simple minded ( @xmath34 ) .
also , effects of centrifugal force was ignored .
similarly , the ratio is independent of the mass accretion rate which should be valid only for low luminosity objects . for high luminosity flows , comptonization would cool the dense region completely ( ct95 ) and
the mass loss will be negligible .
pair plasma supported quasi - spherical shocks forms for low luminosity as well ( kazanas & ellison , 1986 ) .
in reality there would be a dependence on these quantities when full general relativistic considerations of the rotating flows are made .
exact and detailed computations using both the transonic inflow and outflow ( where the compression ratio @xmath39 is also computed self - consistently ) are in progress , and the results would be presented elsewhere ( das & chakrabarti , submitted ) .
figure 2 contains the basic results .
the solid curve shows the ratio @xmath71 as a function of the compression ratio @xmath39 ( plotted from @xmath72 to @xmath73 ) , while the dashed curve shows the same quantity as a function of the polytropic constant @xmath74 ( drawn from @xmath75 to @xmath76 ) , @xmath77 being the adiabatic index .
the solid curve is drawn for any generic compression ratio and the dashed curve is drawn assuming the strong shock limit only : @xmath78 . in both the curves , @xmath79 has been assumed for simplicity . note that if the compression does not take place ( namely , if the denser region does not form ) , then there is no outflow in this model . indeed for ,
@xmath80 , the ratio @xmath71 is zero as expected .
thus the driving force of the outflow is primarily coming from the hot and compressed region . in a relativistic inflow or for a radiation dominated inflow , @xmath81 and @xmath82 . in the strong shock limit , the compression ratio is @xmath83 and the ratio of inflow and outflow rates becomes , @xmath84 for the inflow of a mono - atomic ionized gas @xmath75 and @xmath85 .
the compression ratio is @xmath86 , and the ratio in this case becomes , @xmath87 since @xmath88 is smaller for @xmath85 case , the density at the sonic point in the outflow is much higher ( due to exponential dependence of density on @xmath88 , see , eq .
7 ) which causes the higher outflow rate , even when the actual jump in density in the postshock region , the location of the sonic point and the velocity of the flow at the sonic point are much lower . it is to be noted that generally for @xmath89 shocks are not expected ( c90 ) , but the centrifugal barrier supported dense region would still exist .
as is clear , the entire behavior of the outflow depends only on the compression ratio , @xmath39 and the collimating property of the outflow @xmath90 .
outflows are usually concentrated near the axis , while the inflow is near the equatorial plane .
assuming a half angle of @xmath91 in each case , we obtain , @xmath92 and @xmath93 the ratios of the rates for @xmath82 and @xmath85 are then @xmath94 and @xmath95 respectively .
thus , in quasi - spherical systems , in the case of strong shock limit , the outflow rate is at the most a couple of percent of the inflow .
if this assumption is dropped , then for a cold inflow , the rate could be higher by about fifty percent ( see , fig .
2 ) . it is to be noted that the above expression for the outflow rate is strictly valid if the flow could be kept isothermal at least up to the sonic point . in the event
this assumption is dropped the expression for the outflow rate becomes dependent on several parameters .
as an example , we consider a polytropic outflow of same index @xmath77 but of a different entropy function @xmath96 ( we assume the equation of state to be @xmath97 , with @xmath98 ) .
the expression ( 11b ) would be replaced by , @xmath99 and eq .
( 12 ) would be replaced by , @xmath100 where @xmath101 is the polytropic constant of the flow and @xmath102 and @xmath103 are the adiabatic sound speeds at the starting point and the sonic point of the outflow .
it is easily shown that a power law temperature fall off of the outflow ( @xmath104 ) would yield @xmath105 where , @xmath106 and @xmath107 are the entropy functions of the inflow and the outflow .
this derivation is strictly valid for a non - isothermal flow . since @xmath108 , @xmath109 and @xmath88 , for @xmath110
, @xmath111 is guaranteed provided @xmath112 , i.e. , if the temperature falls for sufficiently rapidly .
for an isothermal flow @xmath113 and the rate tends to be higher
. note that since @xmath114 in this case , any small jump in entropy due to compression will off - balance the the effect of @xmath115 factor . thus @xmath71 remains smaller than unity .
the first factor decreases with entropy jump while the second factor increases with the compression ratio ( @xmath39 ) when @xmath116 .
thus the solution is still expected to be similar to what is shown in fig .
numerical results of the transonic flow using non - isothermal equation of state are discussed in das & chakrabarti ( submitted ) .
although the outflows are common in many astrophysical systems which include compact objects such as black holes and neutron stars , it had been difficult to compute the outflow rates since these objects do not have any intrinsic atmospheres and outflowing matter has to be originated from the inflow only .
we showed in the present paper , that assuming the formation of a dense region around these objects ( as provided by a centrifugal barrier , for instance ) , it is possible to obtain the outflow rate in a compact form with an assumption of isothermality of the outflow at least up to the sonic point and the ratio thus obtained seems to be quite reasonable .
computation of the outflow rate with a non - isothermal outflow explicitly depends on several flow parameters .
our primary goal in this paper was to obtain the rates , and not the process of collimation . since observed jets are generally hollow ( begelman , blandford & rees , 1984 ) they must be externally supported ( either by ambient medium pressure or by magnetic hoop stress ) .
this is assumed here for simplicity .
our assumption of isothermality of the wind till the sonic point is based on ` experience ' borrowed from stellar physics .
momentum deposition from the hot photons from the dense cloud , or magnetic heating may or may not isothermalize the expanding outflow , depending on accretion rates and covering factors .
however , it is clear that since the solid angle at which photons shine on electrons is close to @xmath9 ( as in a narrow funnel wall ) , and since the number of electrons per photon is much smaller in a compact region , it may be easier to maintain the isothermality close to a black hole than near a stellar surface .
the centrifugal pressure supported region that may be present in presence of angular momentum was found to be very useful in explaining the soft and the hard states ( ct95 ; c97 ) , rough agreement with power - law slopes in soft states ( ct95 ; titarchuk , kylafis & mastichiadis , 1997 ) as well as the amplitude and frequency of quasi - periodic oscillations ( molteni , sponholz & chakrabarti 1996 ; ryu , chakrabarti & molteni 1997 ) in black hole candidates .
therefore , our reasonable estimate of the outflow rate from these considerations further supports the view that such regions may be common around compact objects .
particularly interesting is the fact that since the wind here is thermally driven , the outflow rate is higher for hotter gas , i.e. , when comptonization is unimportant , that is , for low accretion rate .
it is obvious that the non - magnetized neutron stars should also have the same dense region we discussed here and all the considerations mentioned here would be equally applicable .
it is to be noted that although the existence of outflows are well known , their rates are not . the only definite candidate whose outflow rate is known with any certainty
is probably ss433 whose mass outflow rate was estimated to be @xmath117 yr@xmath7 ( watson et al .
1986 ) , where @xmath118 is the volume filling factor , @xmath119 is the electron density @xmath10 in units of @xmath120 @xmath121 , @xmath122 is the distance of ss433 in units of @xmath0kpc .
considering a central black hole of mass @xmath123 , the eddington rate is @xmath124 yr@xmath7 and assuming an efficiency of conversion of rest mass into gravitational energy @xmath125 , the critical rate would be roughly @xmath126 yr@xmath7 .
thus , in order to produce the outflow rate mentioned above even with our highest possible estimated @xmath127 ( see , fig .
2 ) , one must have @xmath128 which is very high indeed .
one possible reason why the above rate might have been over - estimated would be that below @xmath129 cm from the central mass ( watson et al . 1986 ) , @xmath130 because of the existence of the dense region at the base of the outflow . in numerical simulations the ratio of the outflow and inflow has been computed in several occasions ( eggum , coroniti & katz , 1985 ; molteni , lanzafame & chakrabarti , 1994 ) .
eggum et al . ( 1985 ) found the ratio to be @xmath131 for a radiation pressure dominated flow .
this is generally comparable with what we found above ( eq .
19a ) . in molteni
et al . ( 1994 ) the centrifugally driven outflowing wind generated a ratio of @xmath132 . here , the angular momentum was present in both inflow as well as outflow , and the shock was not very strong .
thus , the result is again comparable with what we find here .
blandford , r. d. & payne , d. 1982 , mnras , 199 , 883 blandford , r. d. & rees , m.j .
1974 , mnras , 169 , 395 begelman , m.c . &
rees , m.j .
, 1984 , mnras , 206 , 209 begelman , m.c . &
blandford , r.d . & rees , m.j .
1984 , rev .
56 , 255 castor , j. i. , abott , d. c. & klein , r. i. 1975 , , 195 , 157 chakrabarti , s. k. 1984 in active galactic nuclei , ed . j. dyson , ( manchester university press ) chakrabarti , s. k. 1986 , , 303 , 582 chakrabarti , s. k. 1989 , , 347 , 365 ( c89 ) chakrabarti , s. k. 1990 , theory of transonic astrophysical flows ( singapore : world sci . ) ( c90 ) chakrabarti , s. k. 1996a , in accretion processes on black holes , physics reports , v. 266 , no . 5 & 6 , p 229 ( c96a ) chakrabarti , s. k. 1996b , , 464 , 664 ( c96b ) chakrabarti , s. k. 1996c , , 471 , 237 chakrabarti , s. k. 1997 , , 484 , 313 ( c97 ) chakrabarti , s. k. & bhaskaran , p. 1992 , 255 , 255 chakrabarti , s. k. , & titarchuk , l.g .
1995 , , 455 , 623 ( ct95 ) chakrabarti , s. k. , titarchuk , l.g . , kazanas , d. & ebisawa , k. 1996 , a & a supp .
470 , 460 chang , k.m .
& ostriker , j.p .
1985 , , 288 , 428 contopoulos , j. 1995 , , 446 , 67 eggum , g. e. , coroniti , f. v. , katz , j. i. 1985 , , 298 , l41 fukue , j. 1982 , pasj , 34 , 163 hawley , j.w . ,
smarr , l. & wilson , j. 1985 , , 297 , 296 .
kazanas , d. & ellison , d. c. 1986 , 304 , 178 .
knigl , a. 1989 , , 342 , 208 kopp , r.a . & holzer , t.e . , 1976 ,
49 , 43 margon , b. 1984 , araa , 22 , 507 mirabel , i. f. & rodrguez , l. f. 1994 , nat , 371 , 46 molteni , d. , sponholz , h. & chakrabarti , s. k. 1996 , 457 , 805 molteni , d. , lanzafame , g. & chakrabarti , s. k. 1994 , 425 , 161 molteni , d. , ryu . ,
d. & chakrabarti 1996 , , 470 , 460 pauldrach , a. , puls , j. , and kudritzki , r.p .
1986 , astr .
164 , 64 priest , e.r . , 1982 , solar magnetohydrodynamics , ( dordrecht , holland ) ryu , d. , chakrabarti , s. k. , & molteni , d. 1997 , , 474 , 378 tarafdar , s.p .
1988 , , 331 , 932 titarchuk , l. g. 1994 , , 429 , 340 titarchuk , l. g. , mastichiadis , a. , & kylafis , n. g. 1997 , , 487 , 834 watson , m.g . , stewart , g. c. , brinkann , w. , & king , a. r. 1986 , mnras , 222 , 261 fig . 1 : : schematic diagram of inflow and outflow around a compact object
. hot , dense region around the object either due to centrifugal barrier or due to plasma pressure effect or pre - heating , acts like a ` stellar surface ' from which the outflowing wind is developed .
fig . 2 : : ratio @xmath133 of the outflow rate and the inflow rate as a function of the compression ratio of the gas at the dense region boundary ( solid curve ) . also shown in dashed curve is its variation with the polytropic constant @xmath134 in the strong shock limit .
solid angles subtended by the inflow and the outflow are assumed to be comparable . | outflows are common in many astrophysical systems which contain black holes and neutron stars .
difference between stellar outflows and outflows from these systems is that the outflows in these systems have to form out of the inflowing material only .
the inflowing material can form a hot and dense cloud surrounding the compact object either because of centrifugal barrier or a denser barrier due to pair plasma or pre - heating effects .
this barrier behaves like a stellar surface as far as the mass loss is concerned .
we estimate the outflow rate from such considerations .
these estimated rates roughly match with the rates in real observations as well as those obtained from numerical experiments .
= cmmi10 scaled1 = cmmi10 = cmmi7 = cmmib10 scaled1 = cmmib10 = cmmib10 scaled 833 = cmbsy10 scaled1 = cmbsy10 = cmbsy10 scaled 833 1 = 1 = 1 = 9 = 9 = 9 = 10 = 10 = 10 = submitted : 24th july , 1997 ; revised :
6th november , 1998 .
comparison with begelman and blandford work on cauldron model is added . |
physically meaningful interpretation of the measurement of relative motion from the point of view of an accelerated observer in a gravitational field requires the introduction of a special coordinate system ( i.e. fermi coordinates ) along the worldline of the observer . in this coordinate system ,
the equations of relative motion reveal tidal and inertial effects for ultrarelativistic motion ( with speed exceeding the critical value @xmath0 ) that are contrary to newtonian expectations .
while there are general treatments of inertial and tidal effects in fermi coordinates @xcite and the special case of ultrarelativistic motion of particles has been discussed in our recent papers @xcite , the purpose of this work is to present a more systematic and complete description of motion beyond the critical speed @xmath0 . to this end
, the equations of motion in fermi coordinates are discussed in section [ s0 ] and inertial effects are considered in sections [ s2 ] , [ s3 ] and appendix a. section [ s4 ] is devoted to tidal effects .
a summary and brief discussion of our results is contained in section [ s5 ] .
imagine an accelerated observer in a general relativistic spacetime following a worldline @xmath1 , where @xmath2 is the proper time along its trajectory .
the local axes of the observer are given by an orthonormal tetrad frame @xmath3 that is carried along its path according to @xmath4 where @xmath5 is the antisymmetric acceleration tensor , @xmath6 is the local temporal direction and @xmath7 , @xmath8 , form the local spatial triad , and where ( here and throughout this paper ) greek indices run from 0 to 3 , latin indices run from 1 to 3 , the signature of the metric is @xmath9 and @xmath10 , unless specified otherwise . in analogy with the electromagnetic faraday tensor
, we decompose the acceleration tensor into its `` electric '' and `` magnetic '' components @xmath11 ; these are given by the translational acceleration @xmath12 and the rotational frequency @xmath13 , respectively .
that is , the acceleration of the reference trajectory is given by @xmath14 and @xmath15 is the frequency of rotation of the spatial triad with respect to a local nonrotating ( i.e. fermi - walker transported ) triad .
let us next establish a fermi coordinate system @xmath16 in a neighborhood of the reference worldline @xcite .
it turns out that fermi coordinates can be assigned uniquely only to spacetime events that are within a cylindrical region of finite radius along the observer s worldline . at each event @xmath1 on the observer s worldline , consider all spacelike geodesic curves that are normal to this reference worldline at this event .
each event @xmath17 on the resulting hypersurface and within the cylindrical region under consideration is connected to @xmath1 by a unique spacelike geodesic curve that starts at @xmath1 and whose tangent vector @xmath18 at this event is normal to the reference worldline ( that is , @xmath19 ) .
the event @xmath17 is assigned the fermi coordinates @xmath20 , where @xmath21 and @xmath22 here @xmath23 is the proper length of the normal spacelike geodesic segment connecting @xmath1 to @xmath17 .
the fermi coordinate system is admissible in a cylindrical spacetime region around @xmath1 with radius @xmath24 , where @xmath25 is the minimum radius of curvature of spacetime along the reference worldline @xcite .
the observer , at each instant @xmath26 of its proper time `` sees '' a locally euclidean three - dimensional space and `` instantaneously '' determines distances within it using the ( spacelike ) geodesic lengths beginning from its position at the spatial origin of the new fermi coordinates .
a nearby particle worldline that punctures the sequence of three - dimensional spaces at @xmath27 is thus a graph over the reference worldline and each point on the graph is connected to the observer by a `` locally straight '' line ( i.e. a geodesic ) that is normal to the observer s worldline . from the viewpoint of the observer
, the particle has relative coordinate velocity @xmath28 and relative coordinate acceleration @xmath29 .
the general equation of motion for a test particle of mass @xmath30 in the fermi frame is @xmath31 where @xmath32 is the proper time along its worldline , @xmath33 , and @xmath34 is the external force acting on the particle . equation can be written as the system @xmath35 using the identity @xmath36 and the lorentz factor of the particle @xmath37 equation can be expressed as @xmath38 let @xmath39 .
we note that @xmath40 and the physical content of the equation of motion is contained in equation together with @xmath41 and @xmath42 .
these can be written respectively as @xmath43 and @xmath44 it is crucial to recognize that in fermi coordinates the velocity @xmath45 satisfies the condition @xmath46 only at @xmath47 ; indeed , away from the reference trajectory @xmath48 could in principle exceed unity in accordance with the equation of motion .
it is natural to begin our discussion with an accelerated observer in minkowski spacetime . in this case ,
spacelike geodesic segments are straight lines ; therefore , the construction of fermi coordinates is simple .
in fact , equation reduces to @xmath49 . differentiating this relation we find that @xmath50{\rm d}x^0+\lambda^\mu _ { \;\;(i ) } { \rm d}x^i,\ ] ] where @xmath51 and @xmath52 are given by @xmath53 the components @xmath54 of the minkowski metric tensor in fermi coordinates are given by @xmath55 moreover , @xmath56 and the inverse metric is given by @xmath57
the christoffel symbols can be evaluated using and ; the nonzero components are @xmath58_i,\end{aligned}\ ] ] where an overdot denotes differentiation with respect to @xmath59 and @xmath60 is defined by @xmath61 in minkowski spacetime , the fermi coordinates are admissible for @xmath62 .
we note that the structure of the boundary region @xmath63 has been discussed in detail @xcite . in general
, the boundary of the admissible region at a given time @xmath59 is a real quadric cone . if @xmath64 , this surface degenerates into coincident planes .
moreover , if @xmath65 , but @xmath66 , the boundary surface is a hyperbolic cylinder for @xmath67 , a parabolic cylinder for @xmath68 , and an elliptic cylinder for @xmath69 .
as is well known , for @xmath70 , the boundary surface is a circular cylinder of radius @xmath71 .
we now consider the equations of motion in fermi coordinates .
equations and are given by @xmath72 hence , equation has the form @xmath73 where @xmath74 represents the external force per unit mass @xmath75 .\ ] ] it is interesting to note that in equation the purely rotational inertial accelerations are essentially the same as in the nonrelativistic theory @xcite .
inertial accelerations have been discussed by a number of authors @xcite . in particular , it has been shown @xcite that for @xmath76 and @xmath77 , the inertial acceleration experienced by the particle parallel to its motion is given to lowest order in @xmath78 by @xmath79 , where @xmath80 is the unit vector tangent to the spatial path of the particle ; therefore , there is a sign reversal for @xmath81 with consequences that are contrary to newtonian expectations .
note that the acceleration @xmath82 and the inertial acceleration @xmath78 occur in rather different ways in equation .
this difference is a consequence of the absolute character of acceleration in the theory of relativity ; that is , the fact that an observer is accelerated is independent of the choice of coordinates . in particular , while the critical speed associated with @xmath82 is @xmath83 ( on the basis of equation ) , the critical speed associated with @xmath78 in equation is @xmath0 . to obtain the limiting case of lightlike motion ,
we let @xmath84 , where @xmath85 is an affine parameter . in the limit as @xmath86
, we have that @xmath87 ; hence , equation is valid with @xmath88 ( see appendix a ) .
for a rindler observer @xcite ( that is , an observer in hyperbolic motion in minkowski spacetime ) , consider an inertial frame in which the rindler observer is at rest at @xmath89 and has uniform translational acceleration @xmath90 along the @xmath91-axis . the observer s worldline is given by @xmath92 the nonrotating orthonormal tetrad frame along this worldline has the form @xmath93 where @xmath94 and @xmath95 , and the transformation from fermi to inertial coordinates is @xmath96 the rindler coordinates are admissible for @xmath97 and @xmath98 .
we note that the manifold where @xmath99 is the rindler horizon .
it is related to the null cone in the inertial frame , since equations and imply that @xmath100 .
imagine a free particle moving along the @xmath91-axis with speed @xmath101 , @xmath102 , according to @xmath103 from the viewpoint of the rindler observer , this motion is given by @xmath104 for @xmath105 , the particle descends monotonically toward the horizon .
for @xmath106 , its path crosses the path of the rindler observer @xmath107 at two events corresponding to @xmath108 and @xmath109 , where @xmath110 this value corresponds to inertial time @xmath111 , where @xmath112 is the lorentz factor associated with @xmath101 . at either event @xmath113 on the other hand , @xmath114 and @xmath115 .
that is , the free particle ascends from @xmath116 to @xmath117 at @xmath118 such that @xmath119 , then descends back to @xmath116 at @xmath120 and finally approaches the horizon @xmath99 as @xmath121 . for @xmath122 ,
the particle has no acceleration at the crossing events .
we note that for ultrarelativistic motion beyond the critical speed @xmath123 , the inertial acceleration of the free particle has the opposite sign at crossing events compared to intuitive expectations based on newtonian mechanics ; but , other aspects of the motion are not affected .
for instance , the critical speed does not appear to play a role in the proper temporal intervals between the crossing events .
the relevant proper time of the rindler observer @xmath120 given by equation is always less than the proper time of the free particle @xmath124 , since for @xmath125 , @xmath126 that is , the length of the geodesic segment is maximum .
this circumstance , usually called `` the twin paradox , '' is a reflection of the absolute character of acceleration in the theory of relativity . for the limiting case of a null ray ( @xmath127 ) , @xmath128 and @xmath129 .
the ray thus ascends monotonically toward infinity or descends monotonically toward the horizon always with positive acceleration .
moreover , the fermi coordinate speed of the ray can range from zero to infinity .
we now turn our attention to motion in a gravitational field @xcite .
the spacetime metric in fermi coordinates then takes the form @xmath130 where @xmath131 is the projection of the riemann tensor on the tetrad frame of the observer . to simplify matters ,
we concentrate on tidal effects only and assume that the observer follows a geodesic along which a fermi coordinate system is constructed based on a parallel - propagated tetrad frame .
moreover , we neglect external forces on the test particle . in this case , the equations of motion of the free particle in the fermi system , i.e. equations
, reduce to @xmath132 and @xmath133 equality holds for null rays @xmath134 . in this case
the right - hand side of equation is a first integral of the differential equation and higher - order tidal terms can not be neglected .
the fermi coordinate system is admissible within a cylindrical region along the observer s worldline with @xmath135 , where @xmath136 is the supremum of @xmath137 .
equation is the _ geodesic deviation equation _ in fermi coordinates , since it represents the relative motion of the free particle with respect to the fiducial observer that has been assumed here to follow a geodesic . to illustrate the tidal effects for test particles and null rays
, we consider the gravitational field of a kerr black hole .
the kerr metric , in boyer - lindquist coordinates @xmath138 , is given by @xmath139 where @xmath140 and @xmath141 .
the kerr source has mass @xmath142 and angular momentum @xmath143 . for a black hole ,
the specific angular momentum @xmath144 is such that @xmath145 . the observer is taken to be on an escape trajectory along the axis of rotation : it starts at @xmath146 with @xmath147 on this axis and moves radially outward reaching infinity with zero kinetic energy .
its geodesic worldline is given by @xmath148 to determine fermi coordinates , we choose a nonrotating orthonormal tetrad frame along the observer s worldline such that @xmath149 is parallel to the @xmath91-axis . in @xmath138 coordinates , we have @xmath150 where @xmath151 and @xmath152 are given in display . using the axial symmetry of the spacetime , @xmath153 and @xmath154
can be chosen uniquely up to a rotation about the @xmath91-axis .
once a pair is chosen at @xmath146 , then @xmath155 and @xmath156 are parallel propagated along the reference worldline .
we do not require the explicit transformation from boyer - lindquist to fermi coordinates to obtain the curvature components that appear in the approximate equations of motion . indeed , the symmetries of the riemann tensor make it possible to represent these quantities as elements of a symmetric @xmath157 matrix with indices that range over the set @xmath158 .
the result is @xmath159,\ ] ] where @xmath160 and @xmath161 are @xmath162 matrices that are symmetric and traceless and correspond respectively to the electric and magnetic parts of the riemann tensor .
moreover , @xmath163 and @xmath164 , where @xmath165 the curvature components are computed along the worldline of the observer , i.e. at @xmath166 in fermi coordinates .
the electric and magnetic curvatures in and , respectively , only depend upon @xmath167 .
therefore , it suffices to integrate the equation for @xmath168 in to obtain @xmath169 , which when substituted in and with @xmath170 results in @xmath171 and @xmath172 , respectively .
the equation of motion can thus be written as the system @xmath173=0,\\ \label{eq45 } \ddot{y } & -\frac{1}{2}ky \left ( 1 - 2\dot{y}^2+\frac{4}{3 } \dot{x}^2- \frac{2}{3 } \dot{z}^2\right ) + \frac{1}{3 } k\dot{y } ( 5x\dot{x}-7z\dot{z})\nonumber \\ & -q [ \dot{x } \dot{y } \dot{z } y-\dot{z } x ( 1+\dot{y}^2 ) -2\dot{x } z]=0,\\ \label{eq46 } \ddot{z } & + kz \left [ 1 - 2\dot{z } ^2+\frac{1}{3 } ( \dot{x}^2 + \dot{y}^2)\right ] + \frac{2}{3 } k\dot{z } ( x\dot{x } + y\dot{y } ) \nonumber\\ & -q ( x\dot{y } -\dot{x } y)(1-\dot{z}^2)=0,\end{aligned}\ ] ] where we have neglected higher - order tidal terms for the sake of simplicity .
these generalized jacobi equations reduce to the standard jacobi equations when the relative motion is so slow that the velocity terms can be neglected .
consider , for instance , the one - dimensional motion of the test particle along the @xmath91-direction . equations - with @xmath174 for @xmath175 reduce to @xmath176 the behavior of the solutions of this equation , which contains the critical speed @xmath177 , has been studied in detail in our previous work @xcite .
suppose that @xmath178 and @xmath179 ; then , the particle starts from the observer s location and decelerates along the @xmath91-axis toward the critical speed @xmath180 . in this case
@xmath181 given by equation reduces to @xmath182 whose behavior is depicted in figure [ fig:1 ] .
this extends our previous results @xcite and is consistent with the rapid decrease of curvature @xmath183 away from the black hole .
a more interesting situation arises for motion normal to the direction of rotation of the black hole
. equations with @xmath184 for @xmath175 reduce to @xmath185 which turns out to be the general equation for one - dimensional radial motion perpendicular to the rotation axis of the black hole as a consequence of the axial symmetry of the configuration under consideration here .
the corresponding lorentz factor is given by as @xmath186 we note that for ultrarelativistic motion , the particles gain tidal energy and reach the speed of light as in figure [ fig:2 ] for an extreme kerr black hole @xmath187 and in figure [ fig:3 ] for a schwarzschild black hole @xmath188 .
it follows from a comparison of figures [ fig:2 ] and [ fig:3 ] that the tidal acceleration mechanism does not depend sensitively upon the specific angular momentum of the black hole ; in fact , it appears that the time it takes for @xmath189 to diverge is in general slightly longer for @xmath190 as compared to the @xmath191 case .
moreover , the divergence of @xmath189 indicates that our test - particle approach breaks down .
in two recent papers @xcite , we approximated @xmath189 by @xmath192 and plotted its approach toward infinity as a consequence of the ultrarelativistic tidal acceleration mechanism .
the present calculation of @xmath189 is based on an improved approximation scheme and goes beyond our previous work . to reach the speed of light or , equivalently , for the timelike particle worldline to become null the fermi speed @xmath193 should reach a value that is greater than unity as @xmath183 in equation is negative .
we expect that this value of @xmath193 is reached at a later time @xmath59 as demonstrated by a comparison of figures [ fig:2 ] and [ fig:3 ] with our previous results @xcite .
even the present calculations are approximate as they are based on dropping the higher - order tidal terms .
it is clear that at a sufficiently long time @xmath194 , the fermi coordinate system breaks down .
the kinematic breakdown of fermi coordinates , i.e. their inadmissibility , is logically independent of the divergence of @xmath189 , which indicates the failure of the dynamical theory presented here .
our test - particle approximation ignores any back reaction ; furthermore , for charged particles the electromagnetic field configuration near the black hole could enhance or hinder the tidal acceleration mechanism . in any case
, this breakdown of our dynamics nevertheless implies that sufficiently energetic particles emerge from the vicinity of the black hole after having experienced tidal acceleration by the gravitational field of the collapsed source . in this way ,
highly energetic particles may be created by microquasars in our galaxy .
it is important to emphasize that the tidal acceleration mechanism is independent of the horizon structure of the black hole .
these results are interesting in connection with the origin of the highest energy cosmic rays since cosmic rays with energies above the greisen - zatsepin - kuzmin limit @xmath195 are not expected to reach the earth from distant galaxies @xcite .
fermi coordinates constitute a geodesic coordinate system that is a natural extension of the standard cartesian system and is indispensable for the theory of measurement in relativistic physics .
we have discussed the general equation of motion of a pointlike test particle , as well as the limiting case of a null ray , in fermi coordinates .
inertial and tidal effects of ultrarelativistic particles with speeds above the critical speed @xmath196 have been emphasized .
this work goes beyond our previous work in this direction and strengthens the basis for our results and conclusions @xcite .
let us compute the effective external force per unit mass given by equation for the electromagnetic case , namely , @xmath197 in accordance with the lorentz force law for the motion of a particle of charge @xmath198 in an external field @xmath199 . neglecting radiation reaction
, we find that @xmath200 where @xmath85 is the affine parameter defined by @xmath201 .
a massless charged particle does not exist ; therefore , we assume that @xmath202 as @xmath86 . with this assumption
, it turns out that the massless limit of the trajectory of a charged particle is a null geodesic .
there exist nongeodesic null curves in minkowski spacetime .
these may be physically interpreted as follows : imagine the path of a null electromagnetic ray that is reflected from a collection of mirrors fixed in different places in space .
the path consists of straight ( i.e. geodesic ) null segments separated by mirrors .
next imagine a limiting situation involving an infinite number of such idealized pointlike mirrors .
the resulting smooth curve is a nongeodesic null curve .
an example of such a curve is @xmath203 where @xmath204 and @xmath205 are constants .
the generalization to curved spacetime is straightforward .
bm is grateful to donato bini and robert jantzen for helpful correspondence .
xxxxx synge j l 1960 _ relativity : the general theory _ , north - holland , amsterdam marzlin k - p 1994 _ gen . rel .
* 26 * 619 + marzlin k - p 1994 _ phys .
d _ * 50 * 888 + marzlin k - p 1996 _ phys . lett a _ * 215 * 1 mashhoon b 1975 _ astrophys . j. _ * 197 * 705 + mashhoon b 1977 _ astrophys . j. _ * 216 * 591 ni w - t and zimmermann m 1978 _ phys .
* 17 * 1473 + li w - q and ni w - t 1979 _ j. math .
_ * 20 * 1473 + li w - q and ni w - t 1979 _ j. math .
phys . _ * 20 * 1925 chicone c and mashhoon b 2002 _ class .
quantum grav . _
* 19 * 4231 + chicone c and mashhoon b 2004 _ int .
d _ * 13 * 945 chicone c and mashhoon b 2004 _ preprint _ astro - ph/0404170 + chicone c and mashhoon b 2004 _ preprint _ astro - ph/0406005 chicone c and mashhoon b 2004 _ class . quantum grav .
_ , in press mashhoon b 2003 in : _ advances in general relativity and cosmology _ , edited by g. ferrarese , pitagora , bologna , pp . 323334 de felice f 1995 _ phys . rev . a _ * 52 * 3452 estabrook f b and wahlquist h d 1964 _ j. math .
phys . _ * 5 * 1629 defacio b , dennis p w and retzloff d g 1978 _ phys .
d _ * 18 * 2813 + defacio b , dennis p w and retzloff d g 1979 _ phys .
d _ * 20 * 570 jaffe j and shapiro i i 1972 _ phys .
d _ * 6 * 405 jantzen r t , carini p and bini d 1992 _ ann . phys . _
( _ n y _ ) * 215 * 1 + bini d , carini p and jantzen r t 1995 _ class .
quantum grav . _
* 12 * 2549 moliner i , portilla m and vives o 1995 _ phys .
d _ * 52 * 1302 bunchaft f and carneiro s 1998 _ class .
quantum grav . _ * 15 * 1557 rindler w 1969 _ am .
j. phys . _
* 34 * 1174 + rindler w 2001 _ relativity : special , general and cosmological _ , oxford university press , oxford anchordoqui l a , dermer c d and ringwald a 2004 _ preprint _ hep - ph/0403001 | fermi coordinates are the natural generalization of inertial cartesian coordinates to accelerated systems and gravitational fields .
we study the motion of ultrarelativistic particles and light rays in fermi coordinates and investigate inertial and tidal effects beyond the critical speed @xmath0 .
in particular , we discuss the black - hole tidal acceleration mechanism for ultrarelativistic particles in connection with a possible origin for high - energy cosmic rays . |
ultracompact binary systems could provide a promising means to observe directly the influence of gravitational radiation on orbital evolution .
moreover , such systems would be ideal sources for detection with spaced based gravitational radiation observatories ( such as the planned nasa / esa lisa mission ) , opening up the possibility for detailed studies of compact interacting binaries . in recent years a pair of candidate ultracompact systems ; v407 vul ( also known as rx j1914.4 + 2456 ) and rx j0806 + 1527 ( hereafter j0806 ) have been proposed .
these objects were first discovered by rosat ( motch et al .
1996 ; israel et al . 1999 ; beuermann et al .
1999 ) , and initially were suggested to be members of a `` soft '' class of intermediate polars ( ips ) , with the x - ray periods of 569 and 321 s , respectively , representing the putative spin periods of the accreting white dwarfs . since
their discovery extensive follow - up observations have identified the optical counterparts ( ramsay et al .
2000 ; israel et al . 2002
; ramsay , hakala & cropper 2002 ) .
their soft x - ray spectra , the shape of the x - ray modulation , the phasing of the x - ray and optical modulations , the lack of additional longer periods , and the lack of strong optical emission lines have all called into question their ip credentials ( for a discussion see cropper et al .
however , norton , haswell & wynn ( 2004 ) have argued that an ip interpretation is still plausible if the systems are stream - fed , pole - switching accretors ( ie . no accretion disk ) , and are viewed from a nearly face - on geometry .
it was cropper et al .
( 1998 ) who first suggested that v407 vul might be a double - degenerate compact binary .
they proposed a synchronized , magnetic accretor ( polar - like ) model with accretion powering the x - ray flux .
a non - magnetic variant was subsequently proposed by marsh & steeghs ( 2002 ) . in this algol - like model , the accretion stream impacts directly onto a non - magnetic primary , and the spins are not necessarily synchronized with the orbit .
an interesting alternative not requiring accretion was proposed by wu et al .
they suggested a unipolar inductor model , analogous to the jupiter - io system ( clarke et al .
if these systems are indeed compact , and thus the observed x - ray period is the orbital period , then an important discriminating factor is the magnitude and sign of the orbital evolution .
if the systems are accreting stably from degenerate donors , the expected evolution is for the orbit frequency to decrease . in a previous study we ( strohmayer 2002 ) used archival rosat and asca data to explore the evolution of the 1.756 mhz x - ray frequency of v407 vul over an @xmath5 yr time period , and found evidence for a positive frequency derivative , @xmath0 , with
a magnitude consistent with simple expectations for gravitational radiation induced decay of a circular orbit . since a measurement of the frequency evolution places severe constraints on possible models , it is crucial to confirm the initial indications of orbital decay and place tighter constraints on @xmath0 . in this paper
we present the results of new _
observations which confirm an increase in the x - ray frequency and allow us to place much tighter limits on @xmath0 . in
2 we describe the _ chandra _ observations and the data extraction and analysis . in
3 we discuss our phase coherent timing study , and we show that the inclusion of the _ chandra _ data conclusively indicates a positive @xmath6 hz s@xmath3 .
we also discuss the flux variability of the source and excess variance ( timing noise ) in the phase residuals . in
4 we discuss the implications of our findings for the nature of v407 vul .
we conclude in 5 with a brief summary and goals for future observations .
we obtained a total of @xmath7 ksec of exposure on v407 vul with _
chandra _ in february , november , december 2003 , and january , 2004 .
a summary of these observations is given in table 1 .
we observed with acis - s in continuous clocking ( cc ) mode in order to mitigate the effects of pile - up . to maximize the soft photon response the aimpoint was on the backside illuminated ( s3 ) chip . preliminary spectral analysis shows that virtually all the source photons have energies less than 1 kev . for the purposes of our work reported here we used only @xmath8 kev photons .
details of the _ chandra _ spectroscopy will be presented in a future paper . to prepare the data for a precise timing analysis we first corrected the detector read out times to arrival times following the cxc analysis thread on timing with cc mode data .
we then corrected the arrival times to the solar system barycenter using the ciao tool _
we used the same source coordinates employed for analysis of the rosat and asca data ( see strohmayer 2002 ) .
this produces a set of photon arrival times in the barycentric dynamical time system ( tdb ) .
the acis / cc mode produces a one - dimensional `` image '' of the portion of the sky exposed to the detectors .
a realization of this image for the february , 2003 observations is shown in figure 1 .
v407 vul is the strong peak in the `` image . ''
we extracted only events from within the source peak for our timing study .
figure 2 shows a portion of the lightcurve produced from the source extracted events , and demonstrates that _ chandra _ easily detects individual pulses from the source .
we carried out this procedure on all the data and obtained a total of 29,383 _ chandra _ events for our timing analysis .
we performed our coherent timing studies using the @xmath9 statistic .
since the method is described in our earlier paper we do not repeat all the details here ( see strohmayer 2002 ) . as a first step we calculated two @xmath10 power spectra .
the first uses just the earlier rosat and asca data , while the second uses only the new _ chandra _ data .
this procedure allows us to compare the frequencies measured at two widely spaced epochs .
a comparison of the two then determines the nature of any long term trend in the frequency evolution .
this is perhaps the simplest possible analysis that one can do to search for a frequency change .
the results are shown in figures 3 and 4 .
figure 3 shows the @xmath10 power spectra for the two different epochs , rosat + asca ( top ) , and _
chandra _ ( bottom ) . in each case
we plot the difference , @xmath11 , between the peak value of @xmath10 and the values at other frequencies . in the absence of a signal , the @xmath10 statistic has noise properties given by the @xmath12 distribution with @xmath13 degrees of freedom ( dof ) .
the `` forest '' of minima in each plot results from the relatively sparse data sampling .
note , however , that for each dataset the best frequency _ can _ be identified unambiguously ( the one with @xmath14 ) .
this is crucially important , as correctly pointed out by woudt & warner ( 2004 ) , who cautioned that frequency evolution claims for rx j0806.3 + 1527 and v407 vul ( see hakala et al .
2003 ; strohmayer 2003 ; strohmayer 2002 ) might be compromised by an inability to identify the correct frequency at a given epoch . for the rosat + asca data
the next - best alias peak is at @xmath15 from the best frequency .
this is an enormously significant difference in @xmath11 .
the probability that this alias could be the true frequency , compared to a frequency with a @xmath10 value larger by 1300 , is vanishingly small .
the same is true for the _ chandra _ data , for which the nearest alias peak has a @xmath16 .
although a smaller difference than for the rosat + asca data , the probability of this being the true frequency is still incredibly small . as an example , the probability to obtain a difference in @xmath11 of 100.0 ( much smaller than the observed values ) purely by chance is @xmath17 .
having shown that at each epoch we can unambiguously identify the correct frequency , we can now compare these two frequencies and test whether the frequency has changed .
figure 4 shows a zoom - in around the best rosat + asca ( dashed ) and _ chandra _ ( solid ) frequencies , and clearly demonstrates that the frequency is higher in the _ chandra _ epoch .
the increase in frequency is about @xmath18 nhz .
a simple estimate of the mean frequency derivative can be obtained by simply dividing the measured frequency increase by the time interval between the mean epoch of the two datasets .
this gives a value @xmath19 hz s@xmath3 , which is consistent with the limits on @xmath0 deduced from our earlier study as well as the more detailed analysis described below .
we next performed a coherent , phase connected timing analysis to jointly constrain @xmath20 and @xmath0 .
as in our previous study we carried out a grid search by calculating @xmath12 at each @xmath20 , @xmath0 pair ( see strohmayer 2002 for details ) .
the results are summarized in figures 5 , 6 and 7 .
figure 5 shows two pairs of @xmath21 confidence contours .
the large dashed contours represent the constraints derived only from the rosat + asca data , and the solid contours represent the best solution obtained with the inclusion of the new _ chandra _ data .
the analysis definitively favors a positive @xmath22 hz s@xmath3 .
the parameters of our best timing ephemeris are given in table 1 .
although the presence of a long term spin up trend in the data is now indisputable , the minimum @xmath12 value of 390 ( with 98 degrees of freedom ) for our best fit is formally high . because of this we first scaled the @xmath12 values for the whole dataset ( including _ chandra _ ) by the ratio @xmath23 before plotting the confidence contours .
this gives a conservative estimate of the confidence region for @xmath24 and @xmath0 .
figure 6 shows two representations of the phase measurements .
the phase residuals obtained with @xmath25 and our best constant frequency , @xmath20 , for all the data are shown in the top panel . in this representation a positive @xmath0 ( spin - up ) will show up as a quadratic trend with the parabola opening downward .
one can clearly pick out such a trend with the eye .
the phase residuals computed with our best ephemeris ( ie . including the best non - zero value of @xmath0 ) are shown in the bottom panel .
figure 7 is essentially identical to figure 6 except the time gaps have been removed so that all the individual residuals can be more easily seen .
note the much higher statistical quality of the chandra data .
the excess @xmath12 value is indicative of some additional `` timing noise '' in the system . although the rms deviations are qualitatively similar for all the data , the much higher statistical quality of the _ chandra _ data contributes substantially to the excess .
the presence of timing noise could in principle result from fluctuations in the spin up torque , or perhaps from modest changes in the profile of the pulsed emission . by monitoring the character of such fluctuations it should be possible to further constrain models for the spin up torque .
we discuss this further below .
since fluctuations of both sign appear , and over a range of timescales , it seems extremely unlikely that the overall spin up trend could somehow result from a random superposition of fluctuations with just the right amplitude and phasing.using our best timing ephemeris we have calculated folded pulse profiles for the rosat + asca and new _
chandra _ data .
the results are shown in figure 8 .
profile has been displaced vertically by 600 counts for clarity .
phase zero in the plot corresponds to mjd 49257.533373137 ( tdb ) .
the pulse profile appears to be rather stable over the past decade .
an important constraint on models for v407 vul is the present lack of any detectable periodicities with timescales longer than the 1.756 mhz x - ray modulation .
if the system were in fact an ip , then one might expect that the longer , unseen orbital period could eventually be detected with sensitive observations .
the _ chandra _
data are sensitive enough to explore variability on timescales longer than the x - ray pulsation .
we did this by computing a lightcurve using our best fit period as the time bin size .
the resulting lightcurve is shown in figure 9 . in this plot
we have suppressed the time gaps .
individual observations are separated by the vertical dotted lines .
epochs for each of the five observations can be found in table 1 .
the data show evidence for both long term variations in the flux as well as some pulse to pulse variations .
for example , the flux during our next - to - last pointing was a factor of 3 higher than in the initial pointing . the longest pointing ( 2nd panel from the left )
shows evidence for @xmath26 variations in the flux within a few pulse periods .
our 35.4 ksec pointing at v407 vul is the longest contiguous observation of the source to date .
this , combined with the higher signal to noise ratio in the _ chandra _ data compared with rosat or asca , allows , in principle , more sensitive searches for a putative ip orbital period . to search for periods longward of the 1.756 mhz frequency we computed fourier power spectra of each of our observations .
we do not find any robust periodicities which are present in all of our observations . in the power spectrum of the longest pointing a weak sideband structure accompanies the fundamental and first two harmonics of the 1.756 mhz frequency .
the low frequency region of this power spectrum is shown in figure 10 .
the strongest of these peaks are the upper sidebands to the fundamental and first harmonic peaks at 1.756 mhz and 3.512 mhz .
these sidebands are separated in frequency from their parent peaks by 0.4825 mhz , and have leahy normalized powers of 74.8 and 52.5 .
there also appear to be weaker lower sidebands accompanying the first and second harmonic peaks at 3.512 mhz and 5.28 mhz respectively .
these lower sidebands are separated in frequency by 0.273 mhz from their respective harmonic peaks .
the upper sideband frequency separation is close to , but significantly different from , twice the lower sideband separation .
although such a sideband structure could conceivably be intrinsic to the source , and thus could suggest a longer , previously unseen period in the system , we think a more plausible explanation is that these are spurious signals most likely introduced by the dithering of the _ chandra _ pointing position .
_ chandra _ dithers in a lissajous pattern with different periods in the horizontal and vertical directions . for acis ,
the dither periods are 1000 and 707 seconds in the y and z directions , respectively .
indeed , there is a modest peak at a 1000 s period in the power spectrum of our longest observation .
it seems likely that beating between the strong 1.756 mhz modulation and the dither periods could introduce sideband modulations .
moreover , since the strength of power spectral features is proportional to the counting rate , we would expect an astrophysical signal , if robust , to produce a greater signal in our brightest observation , yet there is no evidence for the sideband structure in the observation at the highest counting rate .
based on these arguments we do not think there is yet sufficient evidence to claim detection of a longer period . because of the systematic issues associated with dithering _ chandra _
, it seems likely that xmm observations will provide more definitive results with regard to searches for an unseen orbital period in v407 vul .
our _ chandra _ data definitively shows that the 1.756 mhz x - ray modulation of v407 vul is increasing at a mean rate of @xmath27 hz s@xmath3 .
this rate is consistent with what would be expected if the x - ray modulation is the orbital period of an ultracompact system and it s orbit were decaying via emission of gravitational radiation ( see strohmayer 2002 ) .
based on the observed spin - up there would appear to be two remaining alternatives regarding the nature of the source .
1 ) an ultracompact but detached system powered by electrical energy , as described by wu et al . ( 2002 ) , or 2 ) a low - inclination ip system in which the observed x - ray modulation is effectively the white dwarf spin , and the x - ray flux and observed spin - up are accretion driven ( norton , haswell & wynn 2004 ) .
other ultracompact scenarios appear to be ruled out by the observed spin - up and/or the short orbital period ( see strohmayer 2002 ; cropper et al .
2003 ) . in the ip scenario
the white dwarf accretion is stream - fed and switches from pole to pole and back again on the synodic period .
an accretion disk would form if the magnetospheric radius is less than the circularization radius for the accretion stream . in order not to form a disk and for v407 vul to be stream - fed
the orbital period must satisfy @xmath28 hr ( norton et al .
if this scenario is correct , then the secular spin up is due to accretion onto the white dwarf , and other aspects of the observations , for example , the long term flux variability and the phase timing noise , may be more easily understood in terms of variations in the mass accretion rate .
this identification would also preclude the need to invoke models perceived by some as `` exotic '' ( such as the wu et al .
( 2002 ) electric star scenario )
. however , some intriguing questions remain , for example , the soft x - ray spectrum , the absence ( or at least weakness ) of optical emission lines , and the infrared photometry have all been used to argue against an ip identification ( see cropper et al .
2003 for a discussion ) .
norton et al .
( 2004 ) have argued that the relatively high mass accretion rates , of order @xmath29 yr@xmath3 , required for accretion to power the observed x - ray flux , as well as the nature of stream - fed accretion may mitigate most of these perceived difficulties .
for example , a highly optically thick stream may mask the emission lines , and if the accretion is `` buried '' in the white dwarf surface layers , then it may manifest itself with predominantly soft blackbody emission .
deeper optical spectroscopy would likely provide important new insight .
accretion - induced spin - up should be testable with continued phase coherent monitoring .
comparison of torque fluctuations with x - ray flux ( ie .
mass accretion rate ) could yield a signature of magnetically controlled accretion ( see for example , ghosh & lamb 1979 ) .
moreover , if the frequency derivative changes sign that would strongly support the ip scenario .
our new _ chandra _ data confirm that the 1.756 mhz x - ray frequency of rx v407 vul is increasing at a rate of @xmath30 hz s@xmath3 .
although this rate is consistent with that expected from gravitational radiation losses in a detached ultracompact binary , an ip interpretation in the context of accretion onto a spinning white dwarf can not yet be strongly ruled out . in some ways
the new _ chandra _ results have only deepened the mystery surrounding v407 vul ( and by implication , its sister source , rx j0806.3 + 1527 ) .
after a decade of observations we still do not know the nature of the source with any certainty .
however , a number of future observations could help provide the solutions .
deep pointings with xmm could probe more sensitively for an unseen orbital period , and further timing observations with _
chandra _ will establish constraints on the dependence of torque fluctuations on x - ray flux .
finally , if these do not suffice to crack the mystery , a detection with a spaced - based gravitational radiation observatory ( such as the nasa / esa lisa mission ) would provide definitive evidence for an ultracompact system .
we thank richard mushotzky , zaven arzoumanian , jean swank and craig markwardt for many helpful comments and discussions .
we also thank the referee for his / her detailed comments on the original manuscript , that helped us to significantly improve the paper .
this work made use of data obtained from the high energy astrophysics science archive research center ( heasarc ) at goddard space flight center .
figure 3 : best constant frequency measurements for v407 vul at two different epochs ; the combined rosat + asca data ( top ) , and the new _ chandra _ data ( bottom ) . shown
are the values of @xmath31 as a function of frequency . in this representation
the minima denote the best frequency values for each dataset .
multiple minima are present due to the inclomplete sampling of the datasets . in each case , however , the correct frequency peak can be identified unambiguously ( the peaks centered near @xmath32 ) .
all other peaks are at significantly high values of @xmath33 to be strongly excluded ( see the text for further details ) .
the reference frequency , @xmath20 , is @xmath34 hz .
figure 4 : same as figure 3 but now showing an expanded view of the region around @xmath35 .
the best frequencies from the rosat + asca ( dashed ) and chandra ( solid ) epochs are shown . the frequency measured at the more recent _ chandra _
epoch is clearly higher .
this establishes beyond doubt that the frequency of v407 vul has increased over the last 11 years ( see the text for further details ) .
figure 5 : constraints on @xmath24 and @xmath0 from our phase timing analysis of the rosat , asca and new _
data for v407 vul . shown are the 90 and 99% confidence regions for the rosat + asca data ( dashed ) , and the combined dataset ( rosat , asca and _ chandra _ ) . in plotting the confidence contours for all the data ( solid )
we first scaled the @xmath12 values by the factor 390/98 ( see text for discussion ) .
the inclusion of the _ chandra _ data definitively requires a positive @xmath22 hz s@xmath3 .
figure 6 : phase timing residuals for v407 vul plotted as a function of time using a constant frequency phase model ( ie .
@xmath36 ( top ) , and with the best fitting solution including a positive @xmath37 hz s@xmath3 ( bottom ) . in the top panel
a parabolic shape with the parabola opening downward is indicative of the need for a positive @xmath0 .
such a trend can just about be discerned with the eye .
data are those points beyond year 9 .
figure 7 : same as figure 6 except the time gaps have been removed so that the individual phase residuals can be more easily seen .
the rms deviation of the residuals is also shown ( dotted lines ) .
the _ chandra _
data are those beyond measurment number 58 ( note the smaller error bars ) .
figure 8 : folded pulse profiles for the rosat + asca ( bottom ) and new _ chandra _ data ( top ) using our best timing ephemeris ( see table 1 ) .
profile has been displaced vertically by 600 counts for clarity .
phase zero corresponds to mjd 49257.533373137 ( tdb ) .
figure 9 : flux variability of v407 vul seen with _
chandra _ at five different epochs .
the time binsize was set equal to one cycle of the 1.756 mhz pulsation .
there is clear evidence for large , long term flux modulations as well as shorter timescale fluctuations .
figure 10 : low frequency power spectrum of our longest _ chandra _ pointing .
the power spectrum is normalized such that a pure poisson noise process would have a mean of 2 and be distributed as @xmath12 with two degrees of freedom .
the three peaks that go off - scale are the fundamental , first and second harmonic of the 1.756 mhz modulation .
upper sidebands to the fundamental and first harmonic , and lower sidebands to the first and second harmonics are present .
the horizontal dotted line indicates the power level with a single trial chance probability of @xmath38 .
we suspect the sidebands are an artifact introduced by the _
chandra _ dither ( see the text for a detailed discussion ) .
cccccc + & obsid & instrument & start utc & stop utc & exp ( ksec ) + 1 & 300095 & acis - s ( cc - mode ) & feb 18 , 2003:11:16:30 & feb 18 , 2003:15:23:39 & 13.4 + 2 & 300095 & acis - s ( cc - mode ) & feb 19 , 2003:07:49:17 & feb 19 , 2003:17:58:49 & 35.4 + 3 & 300096 & acis - s ( cc - mode ) & nov 11 , 2003:19:33:05 & nov 11 ,
2003:23:06:22 & 11.2 + 4 & 300111 & acis - s ( cc - mode ) & dec 27 , 2003:02:36:24 & dec 27 , 2003:07:30:14 & 15.4 + 5 & 300112 & acis - s ( cc - mode ) & jan 31 , 2004:21:20:12 & feb 01 , 2004:02:05:30 & 15.1 + | v407 vul ( rx j1914.4 + 2456 ) is a candidate double - degenerate binary with a putative 1.756 mhz ( 9.5 min ) orbital frequency . in a previous timing study using archival rosat and asca data we reported evidence for an increase of this frequency at a rate consistent with expectations for gravitational radiation from a detached ultracompact binary system .
here we report the results of new _
chandra _
timing observations which confirm the previous indications of spin - up of the x - ray frequency , and provide much tighter constraints on the frequency derivative , @xmath0 .
we obtained with _
chandra _ a total of 90 ksec of exposure in two epochs separated in time by 11.5 months .
the total time span of the archival rosat , asca and new _
chandra _
data is now @xmath1 years .
this more than doubles the interval spanned by the rosat and asca data alone , providing much greater sensitivity to a frequency derivative . with the addition of the chandra data an increasing frequency is unavoidable , and the mean
@xmath0 is @xmath2 hz s@xmath3 .
although a long - term spin - up trend is confirmed , there is excess variance in the phase timing residuals , perhaps indicative of shorter timescale torque fluctuations or phase instability associated with the source of the x - ray flux .
power spectral searches for periods longward of the 9.5 minute period do not find any significant modulations , however , the sensitivity of searches in this frequency range are somewhat compromised by the dithering of the chandra attitude .
the observed spin - up is of a magnitude consistent with that expected from gravitational radiation decay , however , the factor of @xmath4 variations in flux combined with the timing noise could conceivably result from accretion - induced spin - up of a white dwarf . continued monitoring to explore correlations of torque with x - ray flux could provide a further test of this hypothesis . |
[ cols="^,^ " , ] for the forward - scattering case of a disk - center observation as a function of the spatial resolution @xmath3 ( in arcseconds ) of a square @xmath4 element .
the dotted lines indicate the resolution for which the average polarization signal is reduced by a factor @xmath5 with respect to the full - resolution case .
, width=302 ] the 3d model atmosphere considered in this paper is a snapshot of a radiation mhd simulation performed with the _ bifrost _ code @xcite taking into account non - equilibrium hydrogen ionization ( see details in * ? ? ?
we used snapshot 385 of the simulation en024048-hion " , which is the same simulation recently made publicly available through the interface region imaging spectrograph project ( iris ; * ? ? ?
* ) at the hinode science data center europe ( http://sdc.uio.no ; see iris technical note 33 ) .
this 3d model atmosphere , which has horizontal periodic boundary conditions , encompasses the upper part of the convection zone , the photosphere , chromosphere , transition region and corona , and is identical to that used by @xcite to investigate the h@xmath0 intensity profile .
the model s magnetic field configuration is representative of an enhanced network region ; it has magnetic field lines connecting two clusters of photospheric magnetic concentrations with two dominant opposite polarity regions 8 mm apart , which reach chromospheric and coronal heights ( see fig .
[ fig : mgf ] ) .
the ly@xmath0 line results from two blended transitions between the ground level of hydrogen , @xmath6 , and the @xmath7 and @xmath8 excited levels . the ly@xmath0 scattering polarization calculations of this paper were carried out fixing the model s neutral hydrogen number density at each spatial grid point ( which were computed with the _ bifrost _ code taking into account time - dependent hydrogen ionization ) and solving the problem of the generation and transfer of polarized radiation assuming statistical equilibrium in a hydrogen model atom with only bound - bound transitions .
the _ bifrost _ ionization balance for hydrogen was computed using the method described in @xcite , which assumes that the lyman transitions are in detailed balance .
the hydrogen model atom we have used in the scattering polarization calculations ( see fig . [ fig : grotr ] ) is composed of the above - mentioned levels and the @xmath9 level . the collisional coupling of this level with the @xmath10-levels may contribute to a slight depolarization of the ly@xmath0 line in regions where the density of perturbers ( protons and electrons ) is significantly larger than @xmath11 @xcite . since in models of the upper solar chromosphere the proton and electron density is typically smaller than @xmath11 , even a three - level model atom ( i.e. , without the @xmath9 level of fig .
[ fig : grotr ] ) could be sufficiently realistic for computing the line - center polarization of ly@xmath0 .
the only level that produces scattering polarization in the core of the ly@xmath0 line is the @xmath12 upper level .
it is interesting to note that the suitability of the above - mentioned three - level model atom implies that the atomic polarization of the @xmath12 level ( population imbalances and quantum coherence between its magnetic sublevels ) is dominated by the absorption of anisotropic radiation in the resonance line transition itself and by the action of the hanle effect on such upper level , without any serious impact of collisional depolarization .
the critical hanle field of the level @xmath12 is @xmath13 g and the scattering polarization of the on - disk ly@xmath0 radiation is sensitive to magnetic strengths between 10 and 100 g , approximately @xcite to compute the emergent stokes profiles @xmath14 , @xmath15 and @xmath16 at each surface point of the atmospheric model we have to solve the following radiative transfer equations for any desired line of sight ( @xmath17 being @xmath14 , @xmath15 or @xmath16 ) @xmath18 where @xmath19 defines the monochromatic optical path length along the ray direction ( with @xmath20 the corresponding geometrical distance ) , @xmath21 is the absorption coefficient , and @xmath22 is the source function of the stokes parameter @xmath17 , with @xmath23 the emission coefficient . is the only non - zero coefficient of the propagation matrix of the stokes - vector transfer equation , because the zeeman splitting is negligible compared to the width of the hydrogen ly@xmath0 line and the angular momentum of its lower level is @xmath24 and , therefore , can not carry atomic alignment . ] while the emission coefficient @xmath25 has contributions from the two blended transitions indicated in fig .
[ fig : grotr ] , @xmath26 and @xmath27 depend only on the atomic polarization of the @xmath8 upper level ( because it is the only ly@xmath0 level that can carry atomic alignment ) .
we quantify the population and atomic polarization of this @xmath28 level by means of the multipolar components of its atomic density matrix ( i.e. , the @xmath29 components , with @xmath30 and @xmath31 , as quantified in the reference system whose quantization axis is along the local vertical ) or , equivalently , by the following quantities ( with the @xmath29 components normalized to the fraction of neutral hydrogen atoms that are in the ground level ) : @xmath32 . in our radiative transfer calculations with porta
we have used the exact @xmath33 , @xmath34 and @xmath35 expressions , which take into account that the two ly@xmath0 transitions are not perfectly overlapped .
such transitions are actually very close in wavelength ( they are only about 6 m apart ) and , for simplicity , we give here the @xmath34 and @xmath35 expressions for the perfect overlapping case : @xmath36 - \sin \chi { \rm im}[{s}^2_1 ] ) \nonumber \\ & - & \frac{\sqrt{3}}{2 } ( 1+\mu^2 ) ( \cos 2\chi { \rm re}[{s}^2_2 ]
-\sin 2\chi \ , { \rm im}[{s}^2_2 ] ) \big{\}}\ ; , \nonumber\end{aligned}\ ] ] and @xmath37+\cos \chi { \rm im}[{s}^2_1 ] ) \label{eq : su } \\ \nonumber & \quad+\mu ( \sin 2\chi \ , { \rm re}[{s}^2_2 ] + \cos 2\chi \ , { \rm im}[{s}^2_2])\big{\}}\;,\end{aligned}\ ] ] where @xmath38 and @xmath39 are the inclination with respect to the solar local vertical and azimuth of the ray , respectively , and @xmath40 ( with @xmath41 and @xmath42 the continuum and the line - integrated opacities , respectively , and @xmath43 the voigt line absorption profile ) . in such equations the reference direction for stokes
@xmath15 is in the plane formed by the ray s propagation direction and the vertical @xmath44-axis .
we point out that eqs .
( [ eq : sq ] ) and ( [ eq : su ] ) , with their corresponding eqs .
( [ eq : rte ] ) , imply that the sources of stokes @xmath15 are the population imbalances ( @xmath45 ) and the quantum coherence ( @xmath46 , with @xmath47 ) , while the stokes @xmath16 signal is produced only by quantum coherence ( @xmath46 , with @xmath47 ) between pairs of sublevels pertaining to the @xmath8 excited level .
to solve eqs .
( [ eq : rte ] ) for any desired line of sight ( los ) we need to know first @xmath21 and the @xmath48 source function components at each spatial grid point . to obtain the self - consistent values of such quantities we have to solve jointly the statistical equilibrium equations for the multipolar components ( @xmath49 ) of the atomic density matrix corresponding to each atomic level @xmath50 and the radiative transfer equations for each of the radiative transitions in the chosen model atom .
the general expressions of such equations can be found in sections 7.2 and 7.13.c of @xcite , which we solved applying porta .
we used a gaussian quadrature for the ray inclinations with respect to the vertical @xmath44-axis ( with 5 inclinations per octant ) and the trapezoidal rule for the ray azimuth ( with 4 azimuths per octant ) . in our radiative transfer calculations we took into account the impact of the doppler shifts , produced by the macroscopic velocity gradients of the 3d atmospheric model , on the anisotropic radiation pumping .
although this physical ingredient is not very important for modeling the scattering polarization of the hydrogen ly@xmath0 line ( because the line is very broad ) , porta takes it fully into account because the dynamic state of the solar atmosphere may produce observable signatures on the scattering polarization of some chromospheric lines @xcite .
our calculations with porta were carried out in the marenostrum iii supercomputer of the barcelona supercomputing center .
the total computing time of the non - lte calculations carried out for the investigation of this paper exceeded one million cpu hours . to grasp better the physics of this complex radiative transfer problem it is useful to write down a simplified set of equations , instead of the general equations for the four - level model atom of fig . [ fig : grotr ]
we have actually solved with porta to obtain all the results shown in this paper . to this end
, we note that the weakly inelastic collisions with protons and electrons that couple the @xmath9 level with the @xmath51 and @xmath52 ones , which we have taken into account in our 3d radiative transfer calculations , do not really have a significant depolarizing effect on the ly@xmath0 polarization in the considered model atmosphere @xcite .
taking also into account that elastic collisions with neutral hydrogen atoms can be safely neglected when modeling the line - core polarization of ly@xmath0 and that the only level contributing to the ly@xmath0 scattering polarization is the @xmath52 one , we have the following approximate expressions for the @xmath53 tensors that quantify the excitation state of the @xmath52 upper level @xcite : @xmath54 \\\\ { { \rm im}[{s}^2_1 } ] \\\\ { { \rm re}[{s}^2_2 } ] \\\\ { { \rm im}[{s}^2_2 } ] \end{array } \right ) & = & { { ( 1-\epsilon)}\over{\sqrt{2 } } } \left ( \begin{array}{r } { j}^2_0 \\\\ { \rm re}[{j}{}^{2}_{1 } ] \\\\ -{\rm im}[{j}{}^{2}_{1 } ] \\\\ { \rm re}[{j}{}^{2}_{2 } ] \\\\
-{\rm im}[{j}{}^{2}_{2 } ] \end{array } \right)\ , \\ & \!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\ !
+ & \!\!\!\!\!\!\!\!\!\!\!\ !
{ \gamma_u}\ , { { ( 1-\epsilon)}}\ , \left ( \begin{array}{cccccc } 0 & m_{12 } & m_{13 } & 0 & 0 \\\\ m_{21 } & 0 & m_{23 } & m_{24 } & m_{25 } \\\\
m_{31 } & m_{32 } & 0 & m_{34 } & m_{35 } \\\\ 0 & m_{42 } & m_{43 } & 0 & m_{45 } \\\\ 0 & m_{52 } & m_{53 } & m_{54 } & 0 \end{array } \right ) \left ( \begin{array}{l } { s^2_0 } \\\\ { { \rm re}[{s}^2_1 } ] \\\\ { { \rm im}[{s}^2_1 } ] \\\\ { { \rm re}[{s}^2_2 } ] \\\\
{ { \rm im}[{s}^2_2 } ] \end{array } \right)\ , , \nonumber\end{aligned}\ ] ] where @xmath55 is the planck function at the ly@xmath0 line - center frequency , @xmath56 is the photon collisional destruction probablility , @xmath57 ( with the magnetic strength @xmath58 in gauss ) , and the @xmath59 coefficients of the magnetic kernel @xmath60 depend on the inclination ( @xmath61 ) of the local magnetic field vector with respect to the vertical @xmath44-axis and on its azimuth ( @xmath62 ) .
the frequency averaged ( over the absorption profile @xmath43 ) radiation field tensors @xmath63 that appear in such equations quantify the mean intensity ( @xmath64 ) , the anisotropy ( @xmath65 ) , and the breaking of the axial symmetry ( the real and imaginary parts of @xmath66 and @xmath67 ) of the incident radiation field at each point within the medium ; their explicit expressions can be found in @xcite , but note that the stokes @xmath14 of the incident ly@xmath0 radiation has to be calculated taking into account that it results from two blended transitions ( see fig .
[ fig : grotr ] ) .
the equations for the @xmath68 ( atomic ) quantities have a clear physical meaning .
in particular , note that the magnetic operator @xmath60 couples locally the @xmath69 components among them ( the hanle effect ) .
the top left panel of fig .
[ fig : jkq ] shows the spatial variation of the kinetic temperature across a 2d slice through the above - mentioned 3d snapshot model atmosphere ; it corresponds to the @xmath44-@xmath70 plane at @xmath71 mm in fig.[fig : mgf ] .
note that the height of the tr changes drastically from one horizontal location to another , outlining a highly crumpled surface . as shown by the white solid curve , the corrugated surface that delineates the model s tr practically coincides with that defined by the ly@xmath0 line - center optical depth unity along the line of sight ( los ) .
the other panels of fig .
[ fig : jkq ] show the radiation field tensors @xmath72 of the ly@xmath0 radiation , normalized to @xmath73 .
such quantities quantify the fractional anisotropy and the symmetry properties of the incident radiation at each point within the medium .
we calculated them after obtaining with the 3d radiative transfer code porta the self - consistent solution in the 3d snapshot model illustrated in fig.[fig : mgf ] , using the four - level model atom of fig .
[ fig : grotr ] and taking into account the ( very weak ) collisional coupling between the @xmath9 and the two @xmath10 levels . of particular interest is the spatial variation of the fractional anisotropy ( top right panel ) ; it is negligible in the photosphere and chromosphere of the model , but it suddenly becomes significant right at the location of the tr with negative values of about 10% .
the other radiation field tensors shown in the remaining panels quantify the breaking of the axial symmetry of the ly@xmath0 radiation at each point within the model atmosphere ; note that at the atmospheric heights where the line center optical depth is unity such tensors may have sizable positive or negative values , depending on the horizontal point under consideration .
the multipolar components of the atomic density matrix ( i.e. , the @xmath29 quantities or , equivalently , the @xmath53 ones ) quantify the excitation state of the atomic level under consideration .
as mentioned above , the scattering polarization of ly@xmath0 is produced by the atomic polarization of its @xmath8 upper level ; therefore , in this section we provide information on the spatial variation of the level s fractional atomic polarization throughout the same @xmath44-@xmath70 plane considered in fig .
[ fig : jkq ] .
( [ eq : mtx-20 ] ) indicate that for @xmath74 ( i.e. , in the unmagnetized case ) each multipolar component @xmath75 of the @xmath8 level is directly proportional to its corresponding radiation field tensor component .
we can say that there is a transfer of ` order ' from the radiation field to the atomic system , and eqs .
( [ eq : rte])([eq : su ] ) show that the atoms of a polarized ( ` ordered ' ) atomic system emit linearly polarized radiation .
( [ eq : mtx-20 ] ) indicate that if all @xmath76 were zero then all @xmath46 would vanish , even in the presence of a magnetic field . through the hanle effect
the magnetic field modifies the polarization of the atomic level according to the @xmath77 term of eqs .
( [ eq : mtx-20 ] ) , and this has an impact on the emergent linear polarization @xmath78 and @xmath79 signals . fig .
[ fig : rhokq ] shows the multipolar components of the density matrix corresponding to the @xmath8 level , normalized to the @xmath80 value of the same level ( which is proportional to its overall population ) .
note that the @xmath81 panel ( which quantifies the population imbalances among the sublevels of the @xmath8 level ) shows mainly negative values at the atmospheric heights where the ly@xmath0 line center optical depth is unity along the los , while each of the quantum coherence shown in the other panels shows both positive and negative values .
this section considers the case of spatially resolved observations .
it shows maps of the line - center scattering polarization , which we have obtained by calculating the emergent stokes profiles of the ly@xmath0 radiation after obtaining the self - consistent solution of the above - mentioned 3d radiative transfer problem .
of particular interest is to investigate the spatial variability of the emergent @xmath78 and @xmath79 line - center signals and their sensitivity to the model s magnetic field . to this end
, we solved the 3d radiative transfer problem of resonance line polarization for the following two cases : ( 1 ) taking into account the hanle effect produced by the magnetic field of the 3d snapshot model atmosphere and ( 2 ) assuming @xmath82 g at each spatial grid point of the 3d model .
we show the results for two line of sights ( los ) defined by @xmath83 ( with @xmath2 the heliocentric angle ) : @xmath84 ( corresponding to a disk center observation ) and @xmath85 ( corresponding to another on - disk observation , but at about 80 seconds of arc from the solar limb ) . in all our calculations of the emergent stokes profiles the azimuth of the corresponding los was @xmath86 that is , the los was contained in the @xmath17-@xmath44 plane . in all the following figures we have chosen the @xmath70-axis as the positive reference direction for stokes @xmath15 , and for any los with @xmath87 we chose this @xmath70-direction as the parallel to the nearest limb .
[ fig : dco ] shows , for the forward scattering case of a disk center observation ( los with @xmath84 ) , the intensity ( top left panel ) , the total fractional linear polarization ( @xmath88 , top right panel ) , and the @xmath78 ( bottom left panel ) and @xmath79 ( bottom right panel ) signals at the center of the emergent ly@xmath0 intensity profile .
the calculated ly@xmath0 line - center intensity variations across the surface of the 3d atmospheric model show fibril - like structures , reminiscent of those observed at 1 arcsecond resolution by the vault sounding rocket ( e.g. , * ? ? ?
although we do not show illustrations here , we point out that the effects of 3d radiative transfer do have an impact on the emergent line - center intensities , as expected from the smoothing effect produced by horizontal radiative transfer under non - lte conditions ( e.g. , * ? ? ?
* ) ; the so - called 1.5d approximation gives a larger line - center intensity contrast across the field of view .
the same applies to other spectral lines such as hydrogen h@xmath0 for which this effect is even more significant due to the lower opacity of the line @xcite .
it is interesting to note that at high spatial resolution the linear polarization signals of the emergent ly@xmath0 line - center radiation are very significant ( see fig . [
fig : dco ] ) . at first sight
this result might appear surprising , given that in fig .
[ fig : dco ] we are considering the forward scattering case of a disk - center observation .
however , in this scattering geometry zero linear polarization should be expected only for the idealized case of an unmagnetized and horizontally homogeneous stellar atmosphere model .
as soon as there is a magnetic field inclined with respect to the symmetry axis of the incident radiation field ( hereafter , inclined field ) and/or horizontal atmospheric inhomogeneities ( e.g. , in the gas kinetic temperature and/or density ) the ensuing symmetry breaking effects can produce significant line scattering polarization signals ( see * ? ? ?
* ; * ? ? ?
* and more references therein ) .
the linear polarization signals of fig .
[ fig : dco ] are due to the breaking of the axial symmetry of the ly@xmath0 radiation at each spatial point within the 3d model atmosphere , as quantified by the real and imaginary parts of the @xmath89 and @xmath90 radiation field tensors , as well to the magnetic field of the atmospheric model ( see eqs . [ eq : mtx][eq : mtx-20 ] ) .
the @xmath78 and @xmath79 line - center signals shown in the bottom panels of fig . [ fig : dco ] have similar values that fluctuate in sign across the field of view , as expected from the particularization of eqs .
( [ eq : sq ] ) and ( [ eq : su ] ) to the case of a los with @xmath84 . in particular , the image corresponding to the total linear polarization at the line center ( see the top right panel of fig .
[ fig : dco ] ) shows an impressive detail of fine - scale structuring .
as explained above , the forward - scattering polarization signals are caused by both the model s horizontal atmospheric inhomogeneities in temperature , density , and macroscopic velocity , and by the magnetic field .
as shown in the top left panel of fig .
[ fig : diff ] , there are very significant forward scattering polarization signals in the absence of magnetic fields .
such zero - field linear polarization signals are solely due to the symmetry breaking caused by the horizontal atmospheric inhomogeneties ; therefore , their very existence indicates that the so - called 1.5d approximation is unsuitable for modeling high spatio - temporal resolution observations of the scattering polarization in ly@xmath0 .
[ fig : diff ] shows that there is a significant difference between the true total fractional linear polarization ( calculated taking into account the hanle effect produced by the magnetic field of the 3d model atmosphere ) and that corresponding to the zero - field reference case .
as expected , the action of the hanle effect in forward scattering geometry is found mainly along the diagonal of the image seen in fig .
[ fig : diff ] , where the magnetization of the model atmosphere is the largest . however , as seen in the figure , the model s magnetic field tends to depolarize the emergent radiation , and this occurs mainly in the strongly polarized regions of the unmagnetized case .
this contrasts with the case of a plane - parallel model atmosphere , where in the absence of magnetic fields the symmetry axis of the incident radiation coincides with the vertical direction and the hanle effect of an inclined magnetic field creates linear polarization in forward scattering geometry but depolarization in close to the limb observations . in our 3d model atmosphere
the symmetry axis of the radiation field that illuminates each of the points of the tr does not in general coincide with the local vertical . in fig .
[ fig : trnorm ] , we attempt to visualize the geometrical complexity of the present 3d problem by showing , at each point of the field of view , the inclination of the vector that is normal to the tr surface . in the unmagnetized case
the strongly polarized regions are typically those for which the inclination of the local normal vector to the tr surface is larger , and depolarization by the hanle effect is statistically more likely . in regions where the local normal vector to the tr surface is predominantly vertical the forward scattering signals of the unmagnetized case are typically lower and the hanle effect produced by the model s magnetic field tends to be more similar to that found in the case of a plane - parallel model atmosphere .
these two situations can be understood by analogy with the scattering polarization signals calculated in a 1d plane - parallel model near the limb and at the disk center , respectively ( cf .
we now consider a los with @xmath85 , which corresponds to the case of an observation relatively close to the solar limb .
[ fig : clo ] shows maps of the line - center intensity ( top left panel ) , of the total fractional linear polarization ( top right panel ) , as well as of the @xmath78 ( bottom left panel ) and @xmath79 ( bottom right panel ) line - center signals .
the particularization of eqs .
( [ eq : sq ] ) and ( [ eq : su ] ) to the los under consideration ( or to any los with @xmath91 ) indicates that @xmath27 ( the emissivity in stokes @xmath16 at each point along the los ) depends only on the quantum coherence ( @xmath92 and @xmath93 ) of the @xmath8 excited level , while @xmath26 ( the emissivity in stokes @xmath15 at each point along the los ) depends also on the population imbalances between the sublevels of the excited level . as indicated by eq .
( [ eq : mtx-20 ] ) , the population imbalances , quantified by @xmath45 , are determined mainly by the radiation field tensor @xmath94 , which at the crumpled surface that delineates the model s tr tends to be negative ( as in the 1d atmospheric models considered by * ? ? ?
we therefore expect that the @xmath78 signals of the emergent ly@xmath0 radiation across the surface of the 3d model have mainly negative sign . for the @xmath79 signals we expect the occurrence of positive and negative values to be equally likely , following the fact ( see eq .
[ eq : mtx-20 ] ) that the radiation field tensors @xmath89 and @xmath90 ( which quantify the symmetry breaking of the incident radiation field ) play a key role on the quantum coherence quantified by @xmath95 and @xmath96 .
the upper panel of fig .
[ fig : histo ] shows the histogram of the line center polarization amplitudes of the emergent ly@xmath0 radiation calculated in forward scattering geometry ( los with @xmath84 ) assuming @xmath82 g .
most of the linear polarization signals have values between 0 and 1% , but there is a clear tail of stronger polarization signals extending up to about 3% . in this forward scattering geometry
the mean fractional linear polarization amplitude at the maximum spatial resolution of the 3d model is @xmath97 the bottom panel of fig .
[ fig : histo ] shows an analogous histogram , but for the case of a los with @xmath85 . as in the @xmath84 case ( see the upper panel of fig .
[ fig : histo ] ) , almost all the surface points are significantly polarized .
in contrast , in this close to the limb scattering geometry there is only a minority of surface points with @xmath98% and there is a significantly larger fraction of points of the field of view showing strong polarization signals with @xmath99% . as a result ,
the mean fractional linear polarization amplitude at the maximum spatial resolution of the 3d model is now @xmath100 .
[ fig : clprofs ] shows , for the case of a close to the limb los with @xmath85 , the synthetic @xmath78 and @xmath79 profiles computed at each @xmath70-point without accounting for any spatial degradation ; the @xmath70-direction here can be assumed to be along the spectrograph s slit , assuming it to be oriented parallel to the nearest limb ( i.e. , the @xmath70 axis ) . of great interest
is to compare the top panels ( the zero - field reference case ) with the bottom panels ( where the hanle effect of the model s magnetic field has been taken into account ) .
in both cases the @xmath78 signals are mainly negative , as expected from the behavior of the radiation field anisotropy ( @xmath101 ) illustrated in fig .
[ fig : jkq ] .
however , the polarization amplitudes are larger in the absence of magnetic fields , with a sizable factor of difference in the atmospheric regions where the magnetic field of the 3d model is relatively strong and inclined ( e.g. , see the region around @xmath102 mm )
. the action of the hanle effect can also be clearly seen in the @xmath79 signals ( right panels ) .
the results shown above indicate that at high spatial resolution we should expect conspicuous linear polarization signals in the ly@xmath0 line core , with amplitudes often larger than @xmath103% .
these @xmath78 and @xmath79 signals are produced by anisotropic radiation pumping in the upper solar chromosphere and they are modified , via the hanle effect , by the vector magnetic field of the model s tr . in the following section ,
we turn our attention to the case of spatially averaged observations .
signal that results from spatially averaging the stokes @xmath14 and @xmath15 line - center signals at each @xmath104 location , averaged over all the los azimuths @xmath39 , with ( solid curve ) and without ( dotted curve ) the hanle effect produced by the model s magnetic field .
the dashed curve corresponds to the hanle effect saturation regime .
the dash - dotted curve shows the calculated clv in the semi - empirical model c of @xcite .
, width=302 ] the figure shows the spatially averaged @xmath78 amplitude vs. the mean magnetic field strength of the model s transition region .
the arrow indicates the @xmath78 signal corresponding to the original 3d model .
, width=302 ] the solar atmosphere is a highly inhomogeneous and dynamic medium , much more complex than the idealization of a 1d , static , plane - parallel semi - empirical model atmosphere . in a 1d model atmosphere with axial symmetry around the local vertical the only possibility to break
the axial symmetry of the incident radiation field at each point within the medium is by the presence of a magnetic field inclined with respect to the solar radius vector through the point under consideration ( see * ? ? ? * and more references therein ) . in the real solar atmosphere
we have , in addition , the symmetry breaking produced by the horizontal inhomogeneities of the atmospheric plasma , in its density , velocity and kinetic temperature .
these ( non - magnetic ) causes of symmetry breaking produce @xmath78 and @xmath79 signals which could be confused with those resulting from the hanle effect . in this paper
we represent the complexity of the tr of the sun by a single snapshot taken from the three - dimensional mhd simulation described in section [ sec : formulation ] .
one possible way to obtain quantitative information on the magnetization of the solar tr is to study the statistical properties of the ly@xmath0 polarization signals , and their relation with the physical quantities that describe the 3d atmospheric model .
the first step towards this goal is to consider the average polarization signals and how they can be exploited to infer information on the average magnetic field strength in the tr . in this section
, we study the sensitivity of the emergent @xmath78 and @xmath79 line - center signals to the model s magnetic field , considering spatially averaged synthetic observations .
as shown below , such spatially - averaged signals can be used to obtain quantitative information on the average magnetization of the solar tr .
if not stated otherwise , the results of this section are shown for a los with azimuth @xmath105 , i.e. , with the @xmath70-axis parallel to the nearest solar limb which is also our reference direction for stokes @xmath15 . in order to simplify the notation , in what follows the spatially averaged quantities @xmath106 , @xmath107 , and @xmath108 will be denoted by @xmath78 , @xmath79 , and @xmath10 .
if the spatial resolution of the observation is reduced , the emergent stokes profiles @xmath14 , @xmath15 , and @xmath16 are averaged over the corresponding resolution element . at the disk center ,
the @xmath78 , @xmath79 , and @xmath10 signals are therefore generally degraded because of cancellation of the positive and negative signals coming from different regions of the atmosphere .
let us consider a square resolution element of size @xmath4 located at the solar disk center . in fig .
[ fig : char ] we show the variation with @xmath3 of the total polarization signal @xmath10 at the ly@xmath0 line center .
this result has been obtained statistically , after computing the emergent stokes profiles for many possible realizations of the resolution element position .
as seen in the figure , at @xmath109 the total polarization line - center signal decreases by a factor two with respect to the maximum resolution value .
three seconds of arc can therefore be regarded as a characteristic spatial scale of the polarized structures seen at the center of the ly@xmath0 line . we can therefore argue that unless the spatial resolution of the observation is significantly better than 3 arcseconds , the polarization of the emergent radiation will be a mixture of signals coming from locations having significantly different plasma structures .
likewise , we can expect that for inferring the average properties of the solar tr , the spatial resolution of a single observation should be much worse than 3 arcseconds , so that the influence of the local inhomogeneities is sufficiently reduced . for the case of a disk - center observation
, the line - center signal @xmath10 tends to zero in the limit of no spatial resolution . for any los with @xmath87
, the spatially averaged signals will be generally non - zero and converging , in the limit of no spatial resolution , to the values given by the center - to - limb variation ( clv ) .
this applies to the unmagnetized case , but as shown in fig .
[ fig : clvav ] it happens also when we take into account the hanle effect produced by the weak magnetic field of the chosen 3d model atmosphere .
the clv of the emergent spectral line radiation contains useful information on the structuring of the solar atmosphere . in quiet regions of the solar surface the observed clv of the stokes-@xmath14 profile of the hydrogen ly@xmath0 line
is known to be negligible ( e.g. , * ? ? ?
however , radiative transfer calculations in the semi - empirical model c of @xcite ( hereafter , fal - c model ) , show that the clv of the @xmath78 line - center signal is rather significant ( cf .
the fig . 2 of * ? ? ? * ) . here
we study the clv of the spatially - averaged @xmath78 line center signals in the chosen 3d model .
the results are shown in fig .
[ fig : clvav ] for the case of azimuthally averaged line of sights of inclination @xmath104 .
only the @xmath78 line - center signal is shown because , as expected , the spatially averaged @xmath79 signal is practically zero for all heliocentric angles .
interestingly , the behavior of the resulting clv of the line - center @xmath78 signal is qualitatively similar to that found in the fal - c model ( see the dash - dotted line in fig .
[ fig : clvav ] ) .
the largest fractional polarization amplitude is found around @xmath110 and it decreases towards larger heliocentric angles .
in contrast to the clv behavior found in 1d models , the decrease of the @xmath78 line - center amplitude when approaching the limb of the 3d atmospheric model is not so significant and in the chosen 3d model the @xmath78 signal is still sizable at near - limb locations . without spatio - temporal resolution ,
the resulting @xmath78 signals drop below the 1% level , with amplitudes similar to those found in 1d semi - empirical models .
therefore , their measurement requires advanced instrumentation , capable of reaching a polarimetric sensitivity of at least 0.1% . via the hanle effect , the spatially - averaged @xmath78 signals are clearly sensitive to the model s magnetic field ( see fig . [
fig : clvav ] ) . following the approach introduced in section 4 of @xcite
, we study the magnetic sensitivity of the spatially - averaged polarization amplitudes in the 3d snapshot model atmosphere .
we define the quantity @xmath111 as the average magnetic field strength at the points where the optical depth at the ly@xmath0 line center is unity for the case of a disk - center observation .
given the height of formation " of the ly@xmath0 line center , @xmath111 can also be understood as the average magnetic field strength in the tr .
we follow the approach of @xcite and multiply the magnetic field strength at each grid point of the 3d model by a dimensionless factor @xmath112 . in this way
, we can study the dependence of the line polarization on the average field strength @xmath113 .
the result for the @xmath78 line center amplitude is shown in fig .
[ fig : bsens ] for a near - limb observation with @xmath85 . is not shown because this quantity remains practically zero for all values of @xmath114 .
] as expected , the maximum polarization signal is found in the unmagnetized case ( @xmath115 ) .
the arrow in the figure indicates the actual value of the average tr magnetization ( the @xmath116 case ) which is @xmath117 g .
the @xmath78 polarization signal gradually decreases as the magnetic field increases .
in contrast to the 2d result shown in fig . 4 of @xcite , in the 3d model the @xmath78 signal is lower and its maximum amplitude is more similar to that found in the fal - c unmagnetized model .
amplitude on the averaging length along the spectrograph s slit , with the slit located at @xmath85 and oriented parallel to the model s solar limb .
the figure shows the polarization signal modified by the hanle effect of the model s magnetic field ( red solid curve ) , the solution without the magnetic field ( dash - dotted curve ) , and the solution obtained with the magnetic field strength increased by a factor 4 at every point of the atmospheric model ( green solid curve ) .
the error bars indicate the standard deviation of the @xmath78 amplitudes calculated from the statistics of randomly positioned slits .
, width=302 ] signal , integrated along the @xmath118 slit , as a function of the averaged magnetic field strength @xmath119 of the model s transition region .
blue curve : slit oriented radially and extending from @xmath120 to @xmath121 .
red curve : slit oriented tangentially with the slit center located at @xmath85 . , width=302 ] the chromospheric ly@xmath0 spectropolarimeter ( clasp ) has been designed for measuring the core of the stokes @xmath14 , @xmath15 and @xmath16 profiles using a spectrograph with a slit of @xmath118 , with a spatial resolution of about @xmath122 , a spectral resolution of 0.1 and a polarimetric sensitivity of @xmath123 at the @xmath124 level ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in order to reach this level of polarimetric sensitivity it will be necessary to significantly sacrifice the temporal resolution and to partially average the observed stokes @xmath14 , @xmath15 , and @xmath16 signals along the direction of the spectrograph s slit , in order to increase the signal - to - noise ratio .
nevertheless , in addition to the important goal of measuring for the first time scattering polarization in a fuv line ( hydrogen ly@xmath0 ) whose core originates in the solar tr , clasp can still aim at constraining the average magnetization of the tr by partial spectrum averaging along the spectrograph slit . an important question to consider
is whether the spectrograph s slit , which selects solar light along a distance of @xmath125 on the solar disk ( i.e. , almost one half of the apparent solar radius ) , should be oriented radially or tangentially with respect to the solar limb , in order to maximize the chances of detection of the expected scattering polarization in ly@xmath0 and the scientific output of the clasp experiment .
we dedicate this section to analyze the pros and cons of both options . although both options seem to be suitable for detecting the scattering polarization signals , we have some reasons to prefer the radial option ( see below ) .
we begin our analysis by considering the case of tangential orientation of the spectrograph s slit , with its center located at the heliocentric angle @xmath85 , i.e. , at an on - disk position where the spatially - averaged @xmath78 signal is maximal ( cf . fig . [
fig : clvav ] ) .
we have studied the dependence of the spatially - averaged @xmath78 line - center signal on the length @xmath126 ( in arc seconds ) along the slit s direction over which we average the computed stokes @xmath14 , @xmath15 and @xmath16 signals .
we have done this analysis by averaging the individual signals corresponding to a large number of randomly positioned virtual slits on top of the 3d model grid .
this has been done by selecting randomly the @xmath127 $ ] point that corresponds to the first point of the slit .
the average is then obtained by integration of the @xmath14 , @xmath15 , and @xmath16 signals along the randomly selected slit .
if the virtual slit crosses the model s boundary , then the integration continues along the same direction but starting at the opposite boundary , with a random position along this boundary .
this way , it is possible to obtain slit - integrated signals for a large number of slits that are longer than the model s dimensions ( which is about @xmath128 ) . we have applied this procedure for averaging lengths between @xmath129 and @xmath130 .
the results are shown in fig .
[ fig : bdet ] for three different degrees of the model s magnetization .
it is important to emphasize that the spatialy - averaged @xmath78 signal depends on @xmath126 : stronger signals are found , on average , for smaller lengths @xmath126 .
this is due to the fact that @xmath78 depends on correlations between @xmath15 and @xmath14 at particular pixels . in our 3d model ,
the regions with lower line intensity are , on average , more polarized than the high - intensity regions . consequently
, stronger signals are obtained for smaller lengths @xmath126 .
the spread of the resulting @xmath78 signals is indicated by the error bars in the figure , which correspond to the standard deviation of each @xmath78 value for any given length @xmath126 . as expected , the larger @xmath126 , the smaller the spread of the values . averaging over the whole @xmath118 distance of the clasp spectrograph s slit would provide a @xmath131 test for detection of the zero magnetization versus the actual model magnetization ( @xmath132 g ) of the 3d model .
similarly , @xmath133 would be sufficient for a test distinguishing between a non - magnetized and a strongly magnetized ( @xmath134 g ) transition region , with a similar level of confidence .
[ fig : bdettara ] shows , both for the radial and tangential slit options , how the spatially - averaged ( including averaging along the whole length of the spectrograph s slit ) @xmath78 line - center signal of the emergent ly@xmath0 radiation from the 3d model varies with the average magnetization value of the model s tr . for each average magnetization , @xmath113 , a very large number of @xmath78 signals were obtained by a method analogous to the one described in the previous paragraph . in the radial - slit option case , the fact that the @xmath78 signal varies with the heliocentric angle ( see fig . [
fig : clvav ] ) has been taken into account .
the amplitude of the @xmath78 signal predicted for the tangential configuration is slightly larger than in the radial case , but the difference is not very important and both slit orientations appear to be suitable for detecting the ly@xmath0 scattering polarization . our reasons in favor of the radial slit option for a sounding rocket experiment such as clasp are the following .
first , in the case of the tangential slit orientation we would have data only for a small interval of @xmath104 values , for example with @xmath135 at the central slit point ( where the polarization amplitude is the largest ) .
obviously , any significant pointing error would compromise the measurement .
in contrast , the radial slit option provides much more information on the clv of the scattering polarization signals , and having such clv data would facilitate reaching more reliable conclusions via detailed confrontations with radiative transfer modeling . note that with a @xmath118 radially - oriented slit one can have information from the solar limb till about @xmath121 . with the radial slit option such clv information would be available both at the line center ( where the hanle effect in ly@xmath0 operates ) and in the nearby wings of the @xmath78 profile ( where effects of partial frequency redistribution and @xmath136-state interference come into play ; see @xcite ) . as shown by these authors , having in addition information on the clv of the @xmath78 wing signals may help to constrain ( together with the observed stokes @xmath14 profiles ) the thermal structure of the upper chromosphere . another reason to prefer the radial slit option
is related with the fact that with the clasp ( which will measure the ly@xmath0 radiation @xmath137 from the line center ) it will be difficult to determine with sufficient precision the zero offset of the linear polarization scale , and having the above - mentioned clv information may help to end up with reliable conclusions in terms of the hanle effect .
clasp will be observing the solar atmosphere for approximately five minutes during which the structure of the dynamical tr will change to some extent . in our numerical experiments , we have only considered an instantaneous state of the solar model atmosphere corresponding to a particular snapshot of the time - dependent simulation .
furthermore , we have assumed an infinitely narrow spectrograph s slit and we have neglected any possible drift of the slit during the simulated observation . because of the fine structuring and dynamical behavior of the solar chromosphere , clasp can not aim at determining the magnetic field at high spatial resolution .
however , the lack of spatio - temporal resolution might actually help us to end up with observables useful to determine the mean field strength of the solar tr .
therefore , the standard deviations shown in figs . [ fig : bdet ] and [ fig : bdettara ] should be understood as being the most pessimistic estimates and the real observation will be of better accuracy .
the outer solar atmosphere is a highly structured and dynamic plasma , much more complex than a uniform , plane - parallel 1d configuration .
very likely , the height of the tr varies from one place in the atmosphere to another , delineating a highly crumpled surface ( e.g. , such as in the top left panel of fig .
[ fig : jkq ] ) .
strong spectral lines originate in this type of spatially complex plasmas , where the horizontal atmospheric inhomogeneities break the axial symmetry of the radiation field .
this symmetry breaking produces changes in the scattering polarization signals , which may compete with those caused by the hanle effect .
therefore , the interpretation of spectropolarimetric observations in chromospheric and tr lines requires solving a complex radiative transfer problem , namely the generation of spectral line polarization in realistic 3d models of the solar atmosphere .
likewise , since the intensity profiles are practically insensitive to the magnetic field vector , the true way to validate or refute the models themselves is by confronting synthesized and measured stokes profiles , because the magnetic field information is encoded in the spectral line polarization . in this paper
we have extended to the full 3d case previous radiative transfer investigations of the scattering polarization in the hydrogen ly@xmath0 line , which were carried out using 1d @xcite and 2d @xcite models of the solar atmosphere .
the 3d model of the extended solar atmosphere chosen for this investigation is representative of the physical conditions of an enhanced network region , with magnetic field lines that reach chromospheric and coronal heights ( see fig .
[ fig : mgf ] ) . using the radiative transfer code porta developed by @xcite we solved the full 3d radiative transfer problem of the generation and transfer of scattering polarization in the hydrogen ly@xmath0 line , and carried out numerical experiments to understand its magnetic sensitivity due to the hanle effect .
the first noteworthy point is that contained in fig .
[ fig : jkq ] , which shows the radiation field tensors that quantify the symmetry properties of the ly@xmath0 radiation .
the fractional anisotropy of the ly@xmath0 radiation ( see the top right panel ) is fully negligible throughout the whole photosphere and chromosphere , but it suddenly becomes significant ( of the order of 10% ) and predominantly negative right at the tr . that the model s tr practically coincides with the corrugated surface where the optical depth is unity along the los . ] the other panels of fig .
[ fig : jkq ] show that the radiation field tensors that quantify the breaking of the axial symmetry of the ly@xmath0 radiation are also negligible below the tr ; i.e. , in the regions where the optical depth at the line center is larger than unity . as the optical depth at the ly@xmath0 line center becomes smaller than unity at the tr , such radiation field tensors become significant . as a result
, the emergent ly@xmath0 radiation shows conspicuous scattering polarization signals , with sizable @xmath78 and @xmath79 line - center amplitudes all over the solar disk , even in the unmagnetized reference case . at full spatial resolution
the most likely fractional polarization amplitude is 1.5% for simulated observations at @xmath85 ( close to the limb scattering geometry ) , while it is 0.7% at @xmath84 ( forward scattering geometry ) .
the polarization signals drop below the 1% level when the spatial resolution is degraded .
for example , for a square resolution element of 3 arc seconds the most likely forward scattering polarization signal decreases by a factor two . in the limit of no spatial resolution
we find that the clv of the line - center @xmath78 signal is qualitatively similar to that found in semi - empirical 1d models of the solar atmosphere , with the largest polarization amplitudes around @xmath85 and with @xmath138 at @xmath84 .
although the average magnetic field strength of the model s tr is only @xmath139 g , the hanle effect in the hydrogen ly@xmath0 line ( whose critical hanle field is 53 g ) produces a significant depolarization ( see fig .
[ fig : bsens ] ) . for stronger magnetic fields
the ensuing depolarization is larger , until it reaches the hanle saturation limit for @xmath140 g . as shown in fig .
[ fig : bdettara ] , for the radial slit case the @xmath141 line - center amplitudes are slightly smaller than in the tangential slit case , but they are equally measurable and show a similarly significant hanle depolarization . in section [ ssec : tara ]
we give our arguments in favor of the radial slit option when observing the ly@xmath0 scattering polarization with the clasp sounding rocket experiment .
another important conclusion is that the hanle effect in forward scattering geometry ( i.e. , when observing the solar disk center ) does not necessarily create linear polarization in the presence of an inclined magnetic field , as it happens in the idealized case of 1d model atmospheres . the fact that in a 3d model of the solar atmosphere the radiation field that illuminates each point of the tr plasma does not have axial symmetry with respect to the local vertical direction implies that even in the absence of a magnetic field forward - scattering processes can produce significant linear polarization signals in ly@xmath0 .
interestingly , we have found and understood that in such a spatially - complex plasma the hanle effect of an inclined magnetic field tends instead to destroy the forward - scattering polarization , as it happens when observing close to the edge of the solar disk where one reaches the case of @xmath142 scattering geometry .
this and the previously summarized results are important for a reliable interpretation of future observations of the ly@xmath0 polarization produced by optically pumped hydrogen atoms in the chromosphere - corona transition region of the sun . financial support by the grant agency of the czech republic through grant and project , as well as by the spanish ministry of economy and competitiveness through projects ( solar magnetism and astrophysical spectropolarimetry ) and
consolider ingenio csd2009 - 00038 ( molecular astrophysics : the herschel and alma era ) is gratefully acknowledged .
this research was also supported by the research council of norway through the grant `` solar atmospheric modelling '' and by the european research council under the european union s seventh framework programme ( fp7/2007 - 2013 ) / erc grant agreement nr .
the radiative transfer simulations presented in this paper were carried out with the marenostrum supercomputer of the barcelona supercomputing center ( national supercomputing center , barcelona , spain ) .
we gratefully acknowledge the computing grants , technical expertise and assistance provided by the barcelona supercomputing center .
we are also grateful to the european union cost action mp1104 ( polarization as a tool to study the solar system and beyond ) for financing short - term scientific missions at the iac that facilitated the development of this research project . | probing the magnetism of the upper solar chromosphere requires measuring and modeling the scattering polarization produced by anisotropic radiation pumping in uv spectral lines .
here we apply porta ( a novel radiative transfer code ) to investigate the hydrogen ly@xmath0 line in a 3d model of the solar atmosphere resulting from a state of the art mhd simulation . at full spatial resolution
the linear polarization signals are very significant all over the solar disk , with a large fraction of the field of view showing line - center amplitudes well above the 1% level . via the hanle effect
the line - center polarization signals are sensitive to the magnetic field of the model s transition region , even when its mean field strength is only 15 g. the breaking of the axial symmetry of the radiation field produces significant forward - scattering polarization in ly@xmath0 , without the need of an inclined magnetic field .
interestingly , the hanle effect tends to decrease such forward - scattering polarization signals in most of the points of the field of view . when the spatial resolution is degraded , the line - center polarization of ly@xmath0 drops below the 1% level , reaching values similar to those previously found in 1d semi - empirical models ( i.e. , up to about 0.5% ) .
the center to limb variation of the spatially - averaged polarization signals is qualitatively similar to that found in 1d models , with the largest line - center amplitudes at @xmath1 ( @xmath2 being the heliocentric angle ) .
these results are important , both for designing the needed space - based instrumentation and for a reliable interpretation of future observations of the ly@xmath0 polarization . |
adiabatic control of a quantum system through its adiabatic evolution path allows robust manipulation and high - fidelity state - preparation @xcite . gradually being recognized as an important requirement in quantum information processing @xcite , it has been under active investigation in recent years @xcite .
the best known examples of the adiabatic control methods are the rapid adiabatic passage ( rap ) @xcite and the stimulated raman adiabatic passage ( stirap ) @xcite . in rap ,
an excitation pulse with a monotonic frequency sweep induces a state vector to follow a population - inversion path in hilbert space traced by an adiabatic state . in stirap ,
a pair of excitation pulses separated in time creates a population - trapping state of a three - level system and the state evolution through the subsequent adiabatic path ensures 100% population transfer from the initial state to the final state without populating the intermediate state .
both of these methods have been widely applied to various quantum systems including atom optics @xcite , nmr @xcite , cavity quantum electrodynamics @xcite , superconducting qubits @xcite , and quantum dots @xcite .
broadband light sources greatly benefit optical approaches to qubit manipulations because of their powerful pulse - shape programming capability @xcite . in ultrafast optics , composing the amplitude and phase of a broadband laser pulse provides various complex pulse shapes , and their usage often plays a crucial role in investigating and engineering new quantum dynamics of atoms and molecules @xcite . of particular
relevance in the context of the present paper is the selective population method of dressed states ( spods ) @xcite which provides a pulse shaping scheme especially in the frequency domain for strong - field controls of multilevel systems . in this paper
we consider a laser pulse shaping method to embed a local nonadiabatic coupling in the middle of a three - level chirped rap process @xcite .
the chirped rap makes a set of cyclic permutation transitions for a three - level system in the @xmath0-type configuration : @xmath4 , @xmath5 , and @xmath6 ( for a positive chirp , and a negative chirp reverses the directions ) , when @xmath7 and @xmath8 are the excited states and @xmath9 is the ground state @xmath9 , and at the first and second adiabatic crossing points the state @xmath9 is interchanged with @xmath7 and @xmath8 , respectively .
so if the transition at the first adiabatic crossing is turned off ( with the new nonadiabatic coupling ) , the states @xmath9 and @xmath7 bypass the crossing and the states @xmath9 and @xmath8 are interchanged at the second crossing and the state @xmath7 is unchanged ( i.e. , @xmath10 , @xmath11 and @xmath6 ) .
we will show that this type of hybrid adiabatic - nonadiabatic interaction can be implemented with a chirped optical pulse with a spectral hole resonant to one of the two excited states . in our method
the laser intensity is used to turn on or off the nonadiabatic coupling , while in a conventional rap approach the spectral chirp sign is used for the selectivity @xcite . using the laser intensity as a control parameter brings along the benefit of spatial beam shaping , which enables simultaneous control of multiple qubits arranged in space ( to be discussed as an application ) .
the remaining sections are organized as follows : we first theoretically study the model hamiltonian for the given shaped - pulse interaction with a three - level system in sec .
ii , where we show that the chirped pulse with a spectral hole can embed the non - adiabatic coupling amid a conventional rap process . after the experimental procedure
is briefly illustrated in sec .
iii , the experimental results are provided in sec .
iv , where the intensity - dependent selectivity of the as - designed hybrid adiabatic - nonadiabatic interaction is presented .
the conclusion follows in sec . v.
the model system is a three - level atom in the @xmath0-type configuration , consisting of two excited energy states , @xmath7 and @xmath8 , and the ground state , @xmath12 ( of respective energies @xmath13 , @xmath14 , and @xmath15 ) .
we consider this system is interacted with a spectrally - shaped laser pulse , a chirped gaussian pulse with a spectral hole , defined in the spectral domain as @xmath16 where @xmath17 and @xmath18 are the center frequencies of the main pulse and the hole , respectively , @xmath19 and @xmath20 are the the bandwidths , and @xmath21 is the chirp parameter @xcite .
the condition @xmath22 $ ] in eq .
makes a complete spectral hole around @xmath23 .
the electric field in the time domain is the inverse fourier transform of @xmath24 , which is given by @xmath25 where @xmath26 and @xmath27 are the chirp parameters for the main and hole pulses , respectively , and @xmath28 is the time shift of the hole with respect to the main pulse . the amplitudes and ( time - independent ) phases of the pulses are respectively given by @xmath29 where @xmath30 is the gaussian pulse width of each chirped pulse @xmath31 .
suppose that the main pulse is frequency - centered between the excited states ( i.e. , @xmath32 ) , with a bandwidth enough to cover the both states ( i.e. , @xmath33 ) and that the hole pulse is resonant to only one of them , say , @xmath7 ( i.e. , @xmath34 and @xmath35 ) . the hamiltonian is then given in the eigenstate basis by @xmath36 - i\hbar r\dot{r}^{-1 } , \label{adiahamiltonian}\end{aligned}\ ] ] where @xmath37 are the eigenstate energies and @xmath38 is the transformation from the bare basis @xmath39 to the eigenstate basis @xmath40 for @xmath41 .
the time evolution of the eigen - energies @xmath42 is plotted in fig .
[ fig1](a ) , where we use @xmath43 , and a positive @xmath44 is chosen to satisfy the minimum hole pulse - width condition for a constant chirp parameter .
the main pulse induces slowly - varying adiabatic passages ( the dotted lines ) and the hole the rapid change of them ( the solid lines ) near the first adiabatic crossing point .
these behaviors are consistent with the temporal profiles of the main pulse and the hole , as shown in fig .
[ fig1](b ) .
note that the the instantaneous frequency of the main pulse becomes equal to @xmath45 at @xmath46 . _
( i ) fully adiabatic coupling regime : _ when the nonadiabatic coupling term , @xmath47 in eq . , is small , the adiabatic condition @xcite is satisfied in all time .
each eigenstate @xmath48 is an adiabatic state , evolving from one bare state @xmath49 to the next one @xmath50 ( cyclically ) , i.e. , @xmath51 up to a global phase .
( a negative chirp reverses the direction of the three - state cyclic permutations . )
so , the result in the fully adiabatic three - state coupling regime is a cyclic transition ( @xmath4 , @xmath52 , @xmath53 ) , similar to the three - level chirped rap @xcite . _
( ii ) hybrid adiabatic - nonadiabatic coupling regime : _ when the adiabatic condition is violated due to the presence of the hole ( we consider the main pulse alone is still adiabatic ) , the given hamiltonian results in a hybrid adiabatic - nonadiabatic transition between @xmath54 and @xmath55 .
since the nonadiabatic coupling is localized in time near @xmath46 , we may consider two separate time regions : @xmath56 and @xmath57 . in the positive time region , the dynamics is a fully adiabatic process , so the state at @xmath58 simply remains till @xmath59 . in the negative time region ,
the eigenstate @xmath60 can be decoupled because it is far - off resonant from @xmath54 .
when we rewrite the hamiltonian in the adiabatic basis of the main pulse ( only ) , the hamiltonian is given under the rotating - wave approximation by @xmath61 \nonumber\\ & - & \frac{\hbar}{2 } r'\left [ \begin{array}{ccc } 0 & \omega_{h}(t ) & 0\\ \omega_{h}^*(t ) & 0 & 0 \\ 0 & 0 & 0 \end{array } \right]r'^{-1 } , \label{2leveladiahamiltonian}\end{aligned}\ ] ] where @xmath62/{2}$ ] are the adiabatic energies , @xmath63 is the detuning ( for the main pulse ) , and @xmath64 is the transform matrix given by @xmath65,\ ] ] with the mixing angle @xmath66 the rabi frequency for the transition from @xmath9 to @xmath7 is defined by @xmath67/\hbar$ ] for each pulse @xmath31 , where the phase factor of the main pulse is added to keep @xmath68 real .
the hamiltonian in eq . can be simplified to be @xmath69\ ] ] with the effective coupling @xmath70 and detuning @xmath71 of the coupled two - level system ( @xmath72 and @xmath73 ) , defined by @xmath74 note that similar coupling and detuning terms are discussed in the context of the zero - area pulse interaction with a two - level system @xcite .
the phase of @xmath70(t ) in eq .
is time - dependent , so the dynamics can be better explained in the interaction - picture .
figure [ fig2](a ) shows the numerical calculation of the coupling @xmath75 and the detuning @xmath76/dt$ ] in the interaction picture .
their plateau region around @xmath46 , the first ( non - adiabatic ) crossing point , manifests a near - resonant two - state coupling , which results in the complete population inversion ( @xmath77 , @xmath78 ) in the adiabatic basis , as shown in fig .
[ fig2](b ) .
when the system evolves further to the second crossing point ( at which the adiabaticity is satisfied ) , the state @xmath55 continues to remain in @xmath55 .
so the given three - level system results in a closed two - level system , @xmath9 and @xmath8 , in the bare - atom basis , plus an isolated state @xmath7 .
figure [ fig2](c ) shows the fully - numerical calculation of the final state populations in the bare - atomic basis using the hamiltonian in eq . which includes the nonadiabatic coupling term , where the calibrated pulse - area @xmath79 is defined with the transform - limited pulse having the same energy of the total pulse .
as the pulse energy increases , the state @xmath80 either remains in @xmath81 ( in the fully non - adiabatic regime for small @xmath79 ) , or evolves to @xmath82 ( in the hybrid adiabatic - nonadiabatic regime for in - between @xmath79 ) , or to @xmath83 ( in the fully adiabatic regime for large @xmath79 ) .
the result indicates the selective transitions to any energy state of the three - level system ( i.e. , @xmath84 , @xmath7 , or @xmath8 ) , controlled with only laser intensity ( in the hybrid adiabatic - nonadiabatic coupling regime ) .
the experimental setup is schematically shown in fig .
[ setup ] .
we used ultrashort optical pulses from a ti : sapphire laser amplifier ( homemade ) operating with a repetition rate of 1 khz and a pulse energy up to 25 @xmath85j .
the spectral bandwidth , the full width at the half maximum ( fwhm ) , was 30 nm ( @xmath86 ) .
the center wavelength was tuned to @xmath87 nm ( @xmath88 ) , the center between the transitions to @xmath89 and @xmath90 energy levels from the ground state @xmath91 of atomic rubidium ( @xmath3rb ) .
each laser pulse was programmed with an acousto - optic pulse shaper ( aopdf , dazzler ) @xcite .
the spectral hole was centered at the transition to @xmath89 , and the linear chirp was varied from @xmath92 to @xmath93 fs@xmath94 by the aopdf .
the shaped laser pulses were focused with an @xmath95 mm lens to the atoms in a vapor cell , and the fluorescence of the atoms induced by the pulses was measured with a ccd ( retiga 3000 ) through a two - lens relay imaging system with @xmath96 mm .
we used two interference bandpass filters centered at 780 nm and 794.7 nm , respectively , to measure the fluorescence from the two excited levels , @xmath89 and @xmath90 .
the filters had a spectral bandwidth of 3 nm and 50 % center transmittance .
, plotted as a function of the hole detuning , @xmath97 , and the calibrated pulse area @xmath79 : ( a ) calculation without the spatial average consideration , ( b ) calculation with the spatial average consideration , and ( c ) experimental results .
the left column ( a1 , b1 , c1 ) corresponds to @xmath98fs@xmath94 ( the minimal hole - pulse width condition ) , the middle column ( a2 , b2 , c2 ) @xmath99fs@xmath94 ( a long hole - pulse ) , and the right column ( a3 , b3 , c3 ) @xmath100fs@xmath94 ( a negatively - chirped pulse ) .
the color scheme of the figures indicates the final ( @xmath59 ) state of the system to be in either @xmath8 ( blue ) or @xmath7 ( red ) .
the calculation used the measured actual laser spectrum , but what used a gaussian spectrum makes little difference . , scaledwidth=90.0% ] figure [ fig4 ]
compares the numerical calculation ( a , b ) with the experimental results ( c ) .
the population difference between the excited states , @xmath101 , is plotted for three chirp parameters : @xmath102 fs@xmath94 ( the minimal hole - pulse condition ) , @xmath103 fs@xmath94 ( a long hole - pulse condition ) , and @xmath104 fs@xmath94 ( a negative chirp ) . the numerical calculation in figs .
[ fig4](a1 ) , [ fig4](a2 ) , and [ fig4](a3 ) shows the chirp - dependent behavior of the given hybrid adiabatic - nonadiabatic interaction . under the minimal hole - pulse condition in fig .
[ fig4](a1 ) , a near - zero detuning ( @xmath105 ) exhibits the as - expected intensity dependence of the selective excitation : as the pulse area ( @xmath79 ) increases , the state evolves to @xmath8 or @xmath7 in the region marked by or , respectively .
near , the system evolves to @xmath8 through the hybrid adiabatic - nonadiabatic interaction .
however , when the adiabatic condition is fully satisfied around , the system evolves to @xmath7 .
note that the region near is the case for a large hole - detuning ; this region exhibits an ordinary chirped - rap behavior , because the far - off - resonant hole plays little role in the overall dynamics .
the long hole - pulse case , in fig . [ fig4](a2 ) , shows an extended non - adiabatic coupling near both the first and second adiabatic crossing points ; thus , the overall dynamics appears to be sensitively dependent on both the hole detuning and the pulse area , as expected . in the negative chirp case , in fig .
[ fig4](a3 ) , the hole pulse is co - located with the second adiabatic crossing point , resulting in , again , an ordinary chirped - rap ( to @xmath8 in this case because of the negative sign of the chirp ) , irrespective of the hole detuning . the second row of fig .
[ fig4 ] is the spatially - averaged calculation of the first row .
because the laser beam has a gaussian spatial profile in the transverse direction , each atom in the atom ensemble interacts with a different laser intensity @xcite .
the gaussian beam wait was 250 @xmath85 m and the rayleigh range ( about 20 cm ) greatly exceeds the size ( about 50 @xmath85 m ) of the imaged area . when this spatial average effect due to the transverse laser beam profile is taken into account , the numerical calculation results in figs .
[ fig4](b1 ) , [ fig4](b2 ) , and [ fig4](b3 ) , showing good agreement with the experimental result in the third row , figs .
[ fig4](c1 ) , [ fig4](c2 ) , and [ fig4](c3 ) , respectively .
finally , the contribution of the hole to the given selective excitation scheme is shown in fig .
[ fig5 ] . by changing the hole depth , defined by @xmath106 in eq . , we plot the degree of inversion defined by @xmath107 , where @xmath108 is the minimal population difference @xmath109 for a given hole depth @xmath106 during the entire evolution .
when the depth is large , @xmath110 $ ] , enough to completely remove the spectrum at @xmath18 , the ground state evolves to the second excited state ( i.e. , @xmath111 ) and @xmath7 is unchanged . on the other hand , no hole ( @xmath112 ) induces the ordinary chirped rap , a cyclic permutation of the energy states . as a function of the hole depth ( @xmath106).,scaledwidth=45.0% ] we now turn our attention to the implications of the results obtained in this study to possible applications .
the first example is the closed adiabatic two - excited state transitions @xmath113 , which can be made by combining the given hybrid interaction and an ordinary chirped rap ( with a negative chirp ) : since an ordinary negatively - chirped rap induces the cyclic state permutations , a subsequent hybrid interaction @xmath111 completes an adiabatic exchange of the excited states , @xmath113 , and @xmath9 is unchanged .
therefore , with this procedure , an ultrafast time - scale adiabatic control among the excited states of a three - level system may be achieved .
the second example is an optical cont rol of @xmath114 qubits arranged in a lattice @xcite . in particular ,
when a short lattice constant makes an conventional optics with focused beams difficult to address individual qubits , our method may provide a solution : our calculation ( not shown ) predicts that spatial beam - shape profiling in conjunction with the given intensity - dependent selective excitation achieves sub - wavelength - scale qubit controls .
for example , the atomic qubit gates constructed based on the rydberg - atom dipole blockade effect often use about 10 @xmath85m - scale optical micro - traps @xcite , so reducing the lattice constant down below one wavelength allows to use significantly lower rydberg energy states , which may be useful for faster quantum gate operations .
in summary , we have studied the hybrid adiabatic - nonadiabatic quantum dynamics of a three - level system in the @xmath0-type configuration , implemented with a chirped laser pulse with a spectral hole .
each adiabatic crossing point of the conventional three - level chirped rap has been found to be individually turned on and off with the spectral hole , enabling selective transition to each excited state by controlling the laser intensity .
experiments have been performed with shaped femtosecond laser pulses and the three lowest energy - levels ( 5s@xmath1 , 5p@xmath1 , and 5p@xmath2 ) of atomic rubidium ( @xmath3rb ) , and the result agrees well with the theoretically analyzed dynamics of the three - level system .
the result indicates that our method , being combined with the ordinary chirped - rap , implements an adiabatic transitions between the two excited states .
furthermore the selective excitations by laser intensity control may have applications including selective excitations of atoms or ions arranged in space in conjunction with laser beam profile programming .
j. beugnon , c. tuchendler , h. marion , a. gaten , y. miroshnychnko , t. r. p. sortais , a. m. lance , m. p. a. jones , g. messin , a. browaeys , and p. grangier , two - dimensional transport and transfer of a single atomic qubit in optical tweezers , nat . phys . * 3 * , 696 ( 2007 ) .
m. j. piotrowicz , m. lichtman , k. maller , g. li , s. zhang , l. isenhower , and m. saffman , two - dimensional lattice of blue - detuned atom traps using a projected gaussian beam array , phys . rev . a * 88 * , 013420 ( 2013 ) .
f. nogrette , h. labuhn , s. ravets , d. barredo , l. bguin , a. vernier , t. lahaye , and a. browaeys , single - atom trapping in holographic 2d arrays of microtraps with arbitrary geometries , phys .
x * 4 * , 021034 ( 2014 ) . | the chirped - pulse interaction in the adiabatic coupling regime induces cyclic permutations of the energy states of a three - level system in the @xmath0-type configuration , which process is known as the three - level chirped rapid adiabatic passage .
here we show that a spectral hole in a chirped pulse can turn on and off one of the two adiabatic crossing points of this process , reducing the system to an effective two - level system .
the given hybrid adiabatic - nonadiabatic transition results in selective excitation of the three - level system , controlled by the laser intensity and spectral position of the hole as well as the sign of the chirp parameter .
experiments are performed with shaped femtosecond laser pulses and the three lowest energy - levels ( 5s@xmath1 , 5p@xmath1 , and 5p@xmath2 ) of atomic rubidium ( @xmath3rb ) , of which the result shows good agreement with the theoretically analyzed dynamics .
the result indicates that our method , being combined with the ordinary chirped - rap , implements an adiabatic transitions between the two excited states .
furthermore the laser intensity - dependent control may have applications including selective excitations of atoms or ions arranged in space when being used in conjunction with laser beam profile programming . |
observations with the euve satellite ( bowyer & malina @xcite ) have provided evidence that a number of clusters of galaxies produce intense euv emission , substantially in excess of the low - energy tail expected from the x - ray emission by the hot , diffuse icm at temperature of a few kev .
euv excesses have been reported for virgo ( lieu et al .
@xcite ) , coma ( lieu et al .
@xcite ) , a1795 ( mittaz , lieu & lockman @xcite ) , a2199 galaxy clusters ( lieu , bonamente & mittaz @xcite ) , a4038 ( bowyer , lieu & mittaz @xcite ) and a4059 ( durret et al .
@xcite ) . however , although there is no doubt about the detection of the euv excess emission for virgo and coma clusters , considerable controversy still exists for the other clusters since euve results are affected by the variation of the telescope sensitivity over the field of view and depend also on subtraction of background signals that do not come from the cluster ( see berghfer & bowyer @xcite and durret et al .
@xcite ) .
the initial explanation for the euv excess emission was that it is produced by a warm ( @xmath2 k ) gas component of the icm ( lieu et al .
@xcite ) . for virgo and coma clusters , bonamente , lieu & mittaz ( @xcite ) have proposed that hi cold clouds might have , at least in part , been released in the intergalactic space by ram pressure stripping . in this framework ,
warm gas can be generated at the interface between the cold phase ( hi clouds ) and the hot icm , e.g. by the _ mixing layer _
mechanism ( fabian @xcite ) .
the resulting scenario is then a three phase gas model , where the hot and warm gas components are responsible for the bulk of x - ray emission and the soft excess emission , respectively , while the cold component accounts for x - ray absorption ( buote @xcite )
. models invoking the presence in galaxy clusters of the warm gas component have problems since they refer to unrelaxed or turbulent gas conditions : the longevity of such multiphase gas would then require the presence of a heat source to carefully balance the radiative cooling , but no obvious heating mechanisms have been found ( fabian @xcite ) . indeed , it is difficult for a warm intracluster gas to remain for a long time at @xmath3k since at this temperature the cooling function for a gas with solar chemical composition has a value @xmath4 erg @xmath5 s@xmath6 ( saito & shigeyama @xcite ) implying for the m87 galaxy warm gas component with metallicity @xmath7 ( gastaldello & molendi @xcite ) and mean number density @xmath8 @xmath9 ( bonamente , lieu & mittaz @xcite ) a cooling time of about 150 myr .
although strong evidence for cooling flows has been found in low - resolution x - ray imaging and spectra , high - resolution observations with the reflection grating spectrometer on the xmm - newton satellite show inconsistencies with the simple cooling flow models .
the main problem is the lack of the emission lines expected from gas cooling below @xmath10 kev . )
, the gas temperature ranges between 1 kev and 2 kev , so that the gas is not cold enough to justify the observed euv emission .
] indeed , observations with both the hopkins ultraviolet telescope ( dixon , hurwitz & ferguson @xcite ) and fuse ( dixonet al .
@xcite ) found no significant far - uv line emission from gas at @xmath11 k. erg @xmath12 s@xmath6 sr@xmath6 for the @xmath13 line and @xmath14 erg @xmath12 s@xmath6 sr@xmath6 for the @xmath15 line .
the absence of a significant far - uv line emission , which is expected from a warm gas component , is also confirmed by fuse observations not only for virgo but also for the coma cluster ( dixon et al .
@xcite ) . ]
also the observation of a large number of clusters by the xmm - newton satellite ( tamura et al .
@xcite ) detected no warm gas component .
all these pieces of evidence seem overwhelming in the sense that a thermal mechanism for the euv excess can be ruled out .
therefore , other mechanisms must be investigated as the source of the euv excess emission in clusters of galaxies .
indeed , inverse compton ( ic ) scattering of cosmic background radiation ( cbr ) photons by relativistic electrons present in the icm was proposed to explain the euv excess in the case of the coma cluster ( hwang @xcite , ensslin & biermann @xcite , ensslin , lieu & biermann @xcite ) . here
, we will examine in detail the case of virgo cluster , referring in particular to the results of a re - analysis of archival euve data for the central region of the cluster performed by berghfer , bowyer & korpela ( @xcite ) . in fig
. 2 of the above cited paper , the azimuthally averaged radial intensity profile of the background - subtracted euv emission in the central region of the virgo cluster ( centered in the m87 galaxy ) is given , as well as the contribution of the low - energy tail of the x - ray emitting hot icm .
comparison between the two curves shows the presence of a diffuse euv excess which extends up to a radius of @xmath16 ( corresponding to @xmath17 kpc , for an assumed m87 distance @xmath18 mpc ) , in the m87 halo region .
( lieu et al .
@xcite ) . ]
the aim of the present paper is to show that ic scattering of cbr and starlight photons by a population ( hereafter denoted as population i ) of mildly relativistic electrons with energy up to a few hundred mev is able to account for the euv excess emission observed from the central region of the virgo cluster , centered in m87 galaxy . as we will see in section 2 , population i electrons have to be described by a power law energy spectrum with an energy cut - off at @xmath19 mev ( or slightly above ) in order to avoid both a hard , non thermal , x - ray excess ( due to ic scattering of cbr photons ) with respect to the upper limit of @xmath20 erg @xmath12 s@xmath6 in the 2 - 10 kev energy band , set by observations ( reynolds et al .
@xcite ) and over - production of radio waves by synchrotron emission ( owen , eilek & kassim @xcite ) .
the existence in the m87 galaxy of a highly relativistic electron population ( which we refer to as population ii ) is well demonstrated by the observations of large - scale radio emission in the 10 mhz - 100 ghz band ( owen , eilek & kassim @xcite ) .
this radio emission is generally interpreted as synchrotron radiation produced by relativistic electrons interacting with the halo magnetic field .
to b exact , taking a value @xmath21 g as typical for the m87 halo magnetic field , one gets an electron energy range @xmath22 gev .
population ii electrons have to be described with a power law energy spectrum with slope @xmath23 in order to fit the observed radio power law spectrum @xmath24 .
electrons belonging to the same population ii , but with energy in the range @xmath25 gev , also could give rise to a hard , non thermal x - ray emission in the 2 - 10 kev region ( through ic scattering of cbr photons ) , but the resulting power is always negligible with respect to the upper limit quoted by reynolds et al .
( @xcite ) .
regarding the origin of population i and ii electrons , it is well known that the main source of non - thermal energy in the m87 galaxy is the active nucleus and jet .
the current bolometric luminosity of the inner active region is @xmath26 erg s@xmath6 - mainly emitted in the radio band ( owen , eilek & kassim @xcite ) - although the kinetic power in the jet ( which matches with the nuclear power for accretion at the bondi rate ) is @xmath27 erg s@xmath6 ( di matteo et al .
this is believed to be a lower limit for the total power available for particle acceleration , since the central engine would have had a cyclic activity with time scale of the order of 100 - 200 myr and today be in a quiescent phase ( owen , eilek & kassim @xcite ; corbin , oneil & rieke @xcite ) .
therefore , simple lifetime considerations indicate that the available current energy content is at least @xmath28 erg . on the other hand , in the past
, the m87 galaxy could have undergone a merging which would have made available a further energy budget ( as suggested for the coma cluster by blasi @xcite ) .
this energy budget could be stored by accelerated protons that , being long - lived in the intergalactic medium ( with lifetimes of the order of a hubble time ) could be confined in the galaxy and then released part of their energy to electrons by coulomb and/or hadronic interactions ( see e.g. ensslin @xcite ) or by shock waves in plasma turbulence .
the observed radio luminosity requires that the power emitted by population ii electrons is @xmath26 erg s@xmath6 . since the cooling time of these electrons
is dominated by synchrotron losses with a time scale of @xmath29 s , one gets a population ii energy budget of @xmath30 erg , much less than the current available energy content .
this implies that most of the total energy made available by the m87 nucleus could be stored in population i electrons .
if these electrons are responsible for the euv excess emission by ic scattering of cbr and starlight photons , their power is about @xmath31 erg s@xmath6 , which multiplied by the ic cooling time @xmath32 s , gives a population i energy content close to that available .
as far as the electron acceleration mechanism is concerned , it is expected that population ii electrons have been directly accelerated in the active nucleus and jet with high lorentz factors . indeed , the two lobes in the observed radio map trace on a large - scale the jet injection , keeping initially the direction of collimated outflows from the core . further diffusion by magnetic field inhomogeneities may give rise to a diffuse population ii component responsible for the overall radio emission .
it is also known that the active nucleus and jet in the m87 inner region are responsible for inflating bubbles of accelerated cosmic rays which rise through the cooling gas at roughly half the sound speed ( churazov et al .
shock acceleration at the bubble boundaries could give rise to population i electrons .
finally we note that the existence for population i electrons of an energy cutoff at @xmath33 250 mev may be justified by considering that a time evolution for power - law spectrum injected electrons is expected as a consequence of transport , diffusion and cooling effects ( see petrosian @xcite for a detailed analysis for the coma cluster ) . in section 2
we calculate the euv source function for the ic scattering by relativistic electrons on cbr and starlight photons and the expected euv excess radial profile towards the m87 galaxy . in section 3
we present our model results and address some conclusions .
as stated in the introduction , we assume that population i electrons account for the large - scale euv excess in m87 galaxy through ic scattering of cbr and starlight photons .
as far as the electron spatial distribution is concerned , despite evidence of granularity in the euv excess map of m87 ( fig . 5 in berghfer ,
bowyer & korpela @xcite ) , here we adopt for simplicity a continuous radial profile as representative of an azimuthally averaged radial distribution of the observed structures .
therefore , the electron density distribution @xmath34 as a function of the electron energy @xmath35 and the radial coordinate @xmath36 , in units of @xmath37 , is @xmath38^{3\beta /2 } } ~g(e_e ) } , \label{diste}\ ] ] where @xmath39 so that the integral number density in the energy band @xmath40 can be calculated as @xmath41 free parameters in eqs .
( [ diste ] ) and ( [ ge ] ) are the constant @xmath42 , the core radius @xmath43 , the exponent @xmath44 of the radial dependence , the electron spectral index @xmath45 , the energy cutoff ( @xmath46 ) and scale ( @xmath47 ) parameters which , in order to avoid over production of hard x - rays and radio waves , turn out to be @xmath48 mev and @xmath49 mev , respectively .
the remaining parameters have to be determined in order to fit the observed euv excess radial profile in the energy band @xmath50 ev . to this end
we make use of the well - known relation giving the ic scattered photon energy ( see e. g. lang @xcite ) @xmath51 in terms of the average energy @xmath52kt@xmath53 of target photons when a black body spectral distribution ( at temperature @xmath54 ) is assumed . in our model the target photons
are supplied by the cbr and starlight ( s ) photons , with temperature @xmath55 k and @xmath56 k , associated with the starlight black body component can be directly obtained from the photometry observations ( see e.g. marcum et al .
@xcite ) towards the m87 galaxy . from these observations
we derive that the absolute magnitudes in the b and v bands are @xmath57 and @xmath58 , respectively .
consequently , the starlight black body temperature turns out to be @xmath59 where @xmath60 is the m87 color index .
] respectively . according to eq .
( [ eqno : eic ] ) , population i electrons in the energy bands @xmath61 mev and @xmath62 mev are involved in order to scatter the cbr and starlight photons , respectively , into the euv energy range @xmath63 ev . the source function for euv photon production through ic scattering
is given by ( lang @xcite ) , in units of @xmath64 , as @xmath65^{3\beta /2 } } \times \left(\frac{e_{\gamma}}{{\rm ev}}\right)^{-\frac{\alpha_e+1}{2 } } } ~ , \end{array } \label{eqno : icsource}\ ] ] where @xmath66 is the target photon number density profile , indicated as @xmath67 and @xmath68 for cbr and starlight photons , respectively .
note that for @xmath69 ev ( corresponding to @xmath70 ) the source function @xmath71 shows a power law spectrum with slope @xmath72 . as far as the average background photon density @xmath66 in eq .
( [ eqno : icsource ] ) is concerned , in the case of cbr photons it is clearly constant ( @xmath73 @xmath9 ) within the m87 galaxy , whereas for starlight photons it is obtained by the relation @xmath74 in which the starlight energy flux @xmath75 , in units of @xmath76 , is obtained from the integration of the local emissivity @xmath77 on the m87 volume @xmath78 we describe the local emissivity by the law ( binney and tremaine @xcite ) @xmath79^ { 3/2 } } ~ , \label{distluce}\ ] ] which implies that the m87 starlight luminosity profile at impact parameter @xmath80 is given by the modified hubble profile @xmath81 } ~. \label{distluce2}\ ] ] in the previous two equations the relation @xmath82 holds ; moreover @xmath83 is connected to the total luminosity through the relation @xmath84}~. \label{distluce3}\ ] ] therefore , the whole photon density profile @xmath85 due to starlight is specified once the total luminosity @xmath86 and the optical core radius @xmath87 are given . according to tenjes , einasto & haud ( @xcite ) and zeilinger , mller & stiavelli ( @xcite ) , we take @xmath88 and @xmath89 pc , respectively . for definiteness
we have taken the m87 galaxy size to be @xmath90 kpc , although our model results depend weakly on the chosen value .
at this point we are ready to estimate the expected euv excess radial profile in the m87 galaxy . by integrating the source function given in eq .
( [ eqno : icsource ] ) along the line of sight at impact parameter @xmath80 ( neglecting internal absorption in m87 ) , we get the surface brightness profile ( in units of @xmath91 ) @xmath92 correspondingly , the total power emitted in the energy band @xmath40 reads @xmath93 the photon number count , in units of @xmath94 , expected to be measured by the deep survey telescope aboard the euve satellite as a function of the angular radius @xmath95 results to be @xmath96 where @xmath97 \simeq [ 50 - 200]$ ] ev is the sensibility energy range of the ds instrument and @xmath98 its effective area given in bowyer et al .
( @xcite ) . in eq .
( [ eqno : counts ] ) we also take into account the galactic absorption through the optical depth @xmath99 , @xmath100 @xmath12 being the milky way column density towards m87 ( hartman & burton @xcite ) and @xmath101 /12 $ ] is accounting for both the cross section of @xmath102 and @xmath103 interaction , as parameterized by morrison & mccammon ( @xcite ) and yan , sadeghpour & dalgarno ( @xcite ) , respectively .
following the formalism discussed in section 2 , we fit the available euv excess data towards the m87 galaxy . for this purpose , experimental points ( given with the respective error bars in fig .
[ fit ] ) are derived from those given in fig . 2 in berghfer , bowyer & korpela ( @xcite ) , by subtracting the contribution of the low - energy tail due to the x - ray emitting icm .
preliminarily , we have fitted the m87 observed euv excess by considering only the ic scattering of population i electrons on cbr photons ( model i , with 11 d.o.f . ) . in this way
, we obtain a poor fit with @xmath104 by considering all the 15 observed points ( see the dotted line in fig .
[ fit ] ) .
value gets reduced to @xmath105 if we do not include the first observed point at @xmath106 . indeed
, a substantial fraction of the observed number count at @xmath106 may be the result of the core and jet activity ( berghfer , bowyer & korpela @xcite ) . ]
consideration of the ic scattering by the same electron population also on starlight photons , distributed accordingly to eqs .
( [ ns ] ) - ( [ distluce3 ] ) , ( model ii , with 9 d.o.f . ) , allows us to substantially increase the expected photon number counts from the innermost part of the galaxy , therefore obtaining a much better confidence level @xmath107 , considering all the 15 points ( see solid line in fig .
[ fit ] ) .
this represents a significant improvement of the fit to all the observed points since the outcome of the f - test implies that model ii has to be preferred over model i within a @xmath108 confidence level ( the f - test value is 3.9 ) .
selected models with @xmath109 give a range of values for @xmath43 and @xmath44 given by @xmath110 kpc and @xmath111 .
as far as the remaining two parameters @xmath45 and @xmath42 - which are actually related by eq .
( [ nde ] ) - our acceptable @xmath109 fits give @xmath112 and , correspondingly , @xmath113 ev@xmath6 @xmath9 .
the obtained relative ratio between the euv luminosity from ic scattering on starlight and cbr photons is in the range of @xmath114 .
the parameter @xmath45 can not be further constrained by euv excess data due to the absence of spectral measurements .
recent experimental results ( reynolds et al .
@xcite ) set an upper limit of @xmath115 erg s@xmath6 @xmath12 for the hard , non - thermal x - ray flux in the 2 - 10 kev energy band from the m87 galaxy .
moreover , radio observations in the 10 mhz - 100 ghz band give a total luminosity @xmath116 erg s@xmath6 ( herbig & readhead @xcite , owen , eilek & kassim @xcite ) .
if we do not introduce an energy cutoff for population i electrons ( see eq .
( [ ge ] ) ) , we would find both a hard , non - thermal x - ray excess emission ( in the 2 - 10 kev energy band , due to ic scattering of cbr photons ) and radio wave excess by synchrotron mechanism , above the upper bound allowed by observations . in the following , as stated in the introduction , we assume that the radio emission is accounted for by population ii electrons described by the same distribution law as in eq .
( [ diste ] ) for population i electrons , but with different parameter values for @xmath117 , @xmath43 , @xmath44 , @xmath45 and without energy cutoff . here
we recall that the characteristic frequency and the differential luminosity of the ultrarelativistic synchrotron radiation are given by ( see e. g. lang @xcite ) @xmath118 and @xmath119 where @xmath120 and @xmath121^{3\beta /2 } } ~4 \pi r^2~dr}~. \label{fe}\ ] ] if we take @xmath122 , so that the predicted radio power law spectrum @xmath123 is in agreement with the observed value , and a value @xmath124 g for the magnetic field strength in the m87 halo ( dennison @xcite ) , there exists a set of acceptable values for the free parameters in eq .
( [ fe ] ) for which the obtained diffuse radio emission from the m87 galaxy in the band 10 mhz 100 ghz is @xmath125 erg s@xmath6 , consistent with the observed value ( herbig & readhead @xcite , owen , eilek & kassim @xcite ) .
it is remarkable that the same set of values for population ii electron parameters that fit radio observations give a hard , non - thermal x - ray flux by ic scattering of cbr photons ( by electrons in the energy range 0.7 - 1 .
7 gev ) in accordance with the upper bounds given by reynolds et al .
( @xcite ) for the m87 galaxy . ) the appropriate population ii electron parameter values . ]
more recently , the xmm - newton observatory has been pointed towards m87 galaxy with the aim of investigating if non - thermal ic emission is present in the arms regions and if it is linked to the power emitted by the radio jet . the upper limit to the non - thermal x - ray emission
has been found to be @xmath126 erg @xmath12 s@xmath6 in the @xmath127 kev band , representing less than @xmath128 of the flux from the thermal components ( belsole et al .
our model results give a hard non - thermal x - ray flux from the inner @xmath129 region of m87 galaxy close to the above upper bound .
clearly , our model results depend on the chosen parameters for population ii electrons which are not , at present , tightly determined by radio observations . for example , by increasing the population ii electron core radius ( or decreasing the value of @xmath44 ) we can dilute the x - ray excess ( and also the radio emission ) within a wider region , leaving the same total power .
a few comments about the energy budget involved for the two electron populations are in order .
observations show that the current bolometric luminosity of the m87 inner active region is @xmath26 erg s@xmath6 - mainly emitted in the radio band ( owen , eilek & kassim @xcite ) - although the kinetic power in the jet is @xmath27 erg s@xmath6 ( di matteo et al .
simple lifetime considerations then indicate that the available electron energy content is at least @xmath130 erg .
however , the total power available for particle acceleration may be larger since the central engine could have had a cyclic activity with a time scale of the order of 100 - 200 myr and today be in a quiescent phase ( owen , eilek & kassim @xcite ; corbin , oneil & rieke @xcite ) .
moreover , in the past the m87 galaxy could have undergone a merging phenomenon which would have made available a further energy budget , initially stored by relativistic protons and later released ( partially ) to electrons by coulomb and/or hadronic interactions or by shock waves in plasma turbulence ( ensslin @xcite ) .
coming back again to our model results , we find that the energy budget associated with the two electron populations is about @xmath131 erg ( mostly in population i electrons ) , which is always less or approximately equal to the total energy associated with thermal electrons .- model with @xmath132 , core radius @xmath133 kpc and temperature @xmath134 kev ( fabricant and gorenstein @xcite ) . moreover , the gas central number density is taken to be @xmath135 @xmath9 so that the total x - ray luminosity in the energy bands 0.2 - 4 kev and 2 - 10 kev turns out to be @xmath136 erg s@xmath6 and @xmath137 erg s@xmath6 , in accordance with the observed values ( forman and jones @xcite ) .
] this result represents an indication that the m87 galaxy is not today in a quiescent ( relaxed ) phase .
however , further more accurate spectroscopic observations in both the euv and x - ray bands ( especially towards the external region of the m87 galaxy ) are necessary in order to confirm the presence of the two electron populations invoked in the present paper .
moreover , observations towards the other galaxy clusters for which the euv emission is still uncertain may establish whether the existence of these two electron populations is a common feature in galaxy clusters or if it is particular to the central , massive , cd galaxies like m87 .
99 andernach h , baker j. r. , van kap - herr a. and wielebinski r. , 1979 , a&a 74 , 93 belsole e. , sauvageot j. l. , bhringer h. , worral d. m. et al . , 2001 ,
a&a 365 , l188 berghfer t. w. , bowyer s. & korpela e. , 2000a , apj 535 , 615 berghfer t. w. , bowyer s. & korpela e. , 2000b , apj 545 , 695 berghfer t. w. & bowyer s. , 2002 , apj 565 , l17 binney j. and tremaine s. , 1987 _ galactic dynamics _ , princeton university press blasi p. 2000 , apj 532 , l9 bonamente m. , lieu r. & mittaz p.d .
, 2001 , apj 546 , 805 bowyer s. & berghfer t.w . , 1998 ,
apj 506 , 502 bowyer s. , lieu r. , lampton m. et al .
, 1994 , apjs 93 , 569 bowyer s. , lieu r. & mittaz j. 1998 , in _ the hot universe _ , ed .
koyama k. , kitamoto s. & itoh m. , dordrecht : kluwer , p. 185
bowyer s. & malina r.f .
, 1991 , _ euv astronomy _
, new york pergamon buote d.a . , 2001 ,
apj 548 , 652 churazov e. , brggen m. , kaiser c. r. , bhringer h. and forman w. , 2001 , apj 554 , 261 corbin m. r. , oneil e. & rieke m. j. , 2002 , aj 124 , 183 dennison b. , 1980 , apj 236 , 761 di matteo t. , allen s. w. , fabian a. c. et al . , astro - ph/0202238 dixon w. , hurwitz m. & ferguson h. c. , 1996 , apj 496 , l77 dixon w. , sallmen s. , hurwitz m. & lieu r. , 2001a , apj 552 , l69 dixon w. , sallmen s. , hurwitz m. & lieu r. , 2001b , apj 550 , l25 durret f. , slezak e. , lieu r. , dos santos s. & bonamente m. 2002 , a&a 390 , 397 ensslin t. & biermann p. , 1998
, a&a 330 , 90 ensslin t. , lieu r. & biermann p. , 1999
, a&a 344 , 409 ensslin t. , 2002 , preprint astro - ph/0206310 fabian a.c . , 1996 , science 271 , 1244 fabian a.c . , 1997 ,
science 275 , 48 fabian a.c . ,
mushotzky r.f .
, nulsen p.e.j . &
peterson j.r . , 2001 ,
mon . not .
r. astron .
321 , l20 fabricant d. & gorenstein p. , 1983
, apj 267 , 535 forman w. & jones c. , 1982 , araa 20 , 547 gastaldello f. & molendi s. , 2002 , apj 572 , 160 hartmann d. & burton w.b . , 1997 , _ atlas of galactic neutral hydrogen _ , cambridge university press herbig t. & readhead a. c. s. , 1992 , apjs 124 , 81 hwang c.y .
, 1997 , science 278 , 1917 lang k.r . , 1974 , _ astrophysical formulae _ , springer - verlag , berlin heidelberg new york lieu r. , mittaz j. p. d. , bowyer s. , et al .
, 1996a , apj 458 , l5 lieu r. , mittaz j. p. d. , bowyer s. , et al . ,
1996b , science 274 , 1335 lieu r. , bonamente m. & mittaz j.p.d . , 1999 , apj 517 , l91 marcum p.m. , oconnel r.w .
, fanelli m.n .
et al . , 2001 ,
apj s 132 , 129 mittaz j.p.d . , lieu r. & lockman f.j . , 1998 ,
apj 498 , l17 molendi s. & pizzolato f. , 2001 , apj 560 , 194 morrison r. & mccammon d. , 1983 , apj 270 , 119 owen f. n. , eilek j. a. & kassim n. e. , 2000 , apj 543 , 611 petrosian v. , 2001 , apj 557 , 560 reynolds c. s. , heinz s. , fabian a. c. & begelman m. 1999 , apj 521 , 99 saito r. & shigeyama t. , 1999 , apj 519 , 48 tamura t. , blecker j. a. m. , kaastra j. s. , ferrigno c. , molendi s. , 2001 , a&a 379 , 107 tamura t. , kaastra j. s. , bleeker j. a. m. , peterson j. r. , 2002 , astro - ph/0209332 tenjes p. , einasto j. and haud u. , 1991 , a&a 248 , 395 yan m. , sadeghpour h.r . & dalgarno a. , 1998 , apj 496 , 1044 zeilinger w.w .
, mller p. & stiavelli m. , 1993 , mon . not
r. astron . | observations with the extreme ultraviolet explorer ( euve ) satellite have shown that the inner region of the virgo cluster ( centered in m87 galaxy ) has a strong extreme ultraviolet ( euv ) emission ( up to @xmath0 ) in excess to the low - energy tail expected from the hot , diffuse intracluster medium ( icm ) .
detailed observations of large scale radio emission and upper limits for hard , non - thermal x - ray emission in the @xmath1 kev energy band have been also reported . here
we show that all available observations can be accounted for by the existence of two electron populations ( indicated as i and ii ) in the m87 galaxy .
the mildly relativistic population i is responsible for the euv excess emission via ic scattering of cbr and starlight photons .
population ii electrons ( with higher energy ) are instead responsible for the radio emission through synchrotron mechanism .
the same electrons also give rise to hard non - thermal x - ray emission ( via ic scattering of cbr photons ) , but the resulting power is always below the upper bounds placed by present observations .
the non - negligible energy budget of the two electron populations with respect to that associated with thermal electrons indicates that the m87 galaxy is not today in a quiescent ( relaxed ) phase .
nuclear activity and merging processes could have made available this energy budget that today is released in the form of relativistic electrons . |
subdwarfs are metal - deficient stars , classically defined as lying below the stellar main - sequence in optical color - magnitude diagrams @xcite .
these objects are in fact not subluminous but rather hotter ( i.e. , bluer in optical colors ) than equivalent mass main - sequence dwarfs , a consequence of their reduced metal opacity @xcite .
cool subdwarfs ( spectral types sdk and sdm ) are typically found to have halo kinematics ( @xmath12 km s@xmath6 ; gizis 1997 ) , and these objects are presumably relics of the early galaxy , with ages of 1015 gyr .
because low mass subdwarfs have lifetimes far in excess of the age of the galaxy , they are important tracers of galactic chemical history , and are representatives of the first generations of star formation . all of the coolest subdwarfs ( [ fe / h ] @xmath130.3 ) and extreme subdwarfs ( [ fe / h ] @xmath140.5 ; gizis 1997 ) currently known have been identified in optical proper motion surveys , most notably luyten s half - second catalog ( * ? ? ?
* lhs ) and two - tenths catalog @xcite , the apm proper motion survey @xcite , and the galactic plane survey of @xcite .
the substantial space velocities of halo subdwarfs allow them to stand out in these surveys amongst the overwhelming multitude of similarly faint , but slowly moving , background stars .
however , only a handful of very cool subdwarfs with spectral types sdm6/esdm6 and later have been found @xcite .
these stars exhibit characteristic spectral signatures of strong metal hydrides ( cah , mgh , alh , feh ) , weak and/or absent metal oxides ( tio , vo , co ) , and enhanced collision - induced ( cia ) h@xmath0 absorption @xcite . with t@xmath15 2900 k @xcite ,
the coolest subdwarfs known have masses just above the hydrogen burning minimum mass ( hbmm ) , which ranges from 0.072 m@xmath9 for solar composition ( @xmath16 ) to 0.092 m@xmath9 for @xmath17 @xcite .
in contrast , hundreds of significantly cooler population i , or disk , dwarfs have been identified , predominately in the wide - field optical and near - infrared ( nir ) surveys denis @xcite , 2mass @xcite , and sdss @xcite .
these discoveries include many dozens of substellar objects , and extend well beyond the m spectral class into the l @xcite and the t @xcite classes .
the success of recent surveys in identifying very cool dwarf star and brown dwarfs derives from their use of red - sensitive ccds and infrared array detectors .
as we proceed to cooler temperatures , the majority of emitted flux from both disk dwarfs and subdwarfs moves to nir wavelengths . unfortunately ,
proper motion surveys of sufficient temporal breadth to identify halo stars rely primarily on optical , typically photographic , imaging .
factoring in their intrinsic rarity ( 0.3% number density compared to disk dwarfs ; reid & hawley 2000 ) and substantial evolution toward lower luminosities ( * ? ? ?
* ; * ? ? ?
* see figure 3 ) , it is clear why few very cool subdwarfs have been found . in this article
, we present the discovery of a new cool subdwarf , identified in the 2mass database .
this object , 2mass j05325346 + 8246465ss@xmath5ddmmss[.]s '' .
the suffix conforms to iau nomenclature convention and is the sexigesimal r. a. and decl . at j2000 equinox .
we adopt a shorthand notation of `` 2mass hhmm@xmath5ddhh '' throughout the remainder of this letter .
] , is a high motion source with a spectral morphology consistent with a metal deficient l dwarf . in @xmath18 2
we discuss the identification of 2mass 0532 + 8246 and subsequent nir and optical observations obtained at the palomar 60 telescope . in @xmath18 3
we describe spectroscopic observations of 2mass 0532 + 8246 obtained with the near - ir spectrometer ( * ? ? ?
* ; * ? ? ?
* hereafter nirspec ) and low resolution imaging spectrograph ( * ? ? ?
* hereafter lris ) at keck observatory . in @xmath18 4
we derive the space motion of this object from the imaging and spectroscopic data . in @xmath18 5
we discuss the substellarity of 2mass 0532 + 8246 based on subsolar metallicity models from @xcite .
we discuss our results in @xmath18 6 .
2mass 0532 + 8246 was identified in a search for t dwarfs using the 2mass working database @xcite .
it was selected for its blue nir colors ( @xmath1 ) and lack of an optical counterpart at the 2mass position in both poss - i and poss - ii photographic plates .
a finder chart is given in figure 1 , and astrometric and photometric data from 2mass are listed in table 1 .
follow - up nir imaging using the palomar 60 infrared camera ( * ? ? ?
* hereafter ircam ) on 25 september 1999 ( ut ) determined that this object is not an uncatalogued minor planet .
gunn @xmath19- and @xmath20-band images obtained with the palomar 60 facility ccd camera on 18 november 2002 and 25 january 2003 ( ut ) further confirmed the red optical / nir colors of this object .
aperture photometry from 18 november 2002 , calibrated with observations of the sdss photometric standard bd + 33@xmath21 4737 ( * ? ? ?
* correcting ( @xmath22 ) @xmath23 ( gunn @xmath19 , gunn @xmath20 ) using photometry from kent 1985 ) yield @xmath24 ( 5@xmath25 upper limit ) and @xmath26 . as the nir and optical / nir colors of this object
are consistent with those of other identified late - type l and t dwarfs @xcite , 2mass 0532 + 8246 was initially believed to be a firm t dwarf candidate .
we obtained nir spectra of 2mass 0532 + 8246 on 24 december 2002 ( ut ) using nirspec on the keck - ii 10 m telescope .
conditions were clear with seeing of 0@xmath46 .
we observed the object in three spectral configurations , n3 ( 1.14 1.38 @xmath27 ) , n5 ( 1.51 1.79 @xmath27 ) , and n6b ( 1.94 2.32 @xmath27 ) , corresponding roughly to the j , h , and k nir bands .
spectral resolution was r @xmath28 2000 for our 0@xmath438 slit . for each observation ,
two exposures of 300s each were obtained , dithering 20@xmath29 along the slit .
all observations were made in the airmass range 2.63.0 . immediately after each sequence of exposures , internal flat field and
near arc lamps were observed for calibration , and the a0 star hd 34360 was observed just before or after the target source in each spectral order at similar airmasses .
data were reduced using the redspec package ; a detailed description of our reduction procedures is given in @xcite .
final flux calibration of the individual spectral orders was done using 2mass photometry ; however , because the orders do not span the entire photometric bandpasses @xcite , our final flux calibration may be somewhat underestimated at j - band , but reasonably correct for the h- and k - bands . the resulting nir spectrum is shown in figure 2a .
the 1.52.4 @xmath27 spectral energy distribution is fairly smooth and blue , reminiscent of an early - type main - sequence star or white dwarf , and quite unlike the much redder nir spectra of late - type m and l dwarfs ( cf .
overlay of the l7 denis 0205@xmath21159ab in figure 2a ; delfosse et al .
1997 ; kirkpatrick et al . 1999 ; mclean et al .
2003 ) . however , the red optical / nir colors and optical molecular bands ( see below ) of 2mass 0532 + 8246 are inconsistent with a hot stellar atmosphere .
we instead attribute the shape of the nir spectrum to enhanced absorption by the cia h@xmath0 1 - 0 quadrupole band centered near 2.5 @xmath27 @xcite .
this pressure - sensitive band , present in late - type l and t dwarfs @xcite and cool subdwarfs @xcite , is generally broad and featureless , consistent with the observed spectrum .
strong h@xmath0 absorption may explain the absence of ch@xmath30 absorption bands at 1.6 and 2.2 @xmath27 , the defining features of t dwarfs ; and the 2.3 @xmath27 co band , a key spectral feature in m , l , and early - type t dwarf spectra .
on the other hand , h@xmath0o absorption appears to be present at 1.31.5 and 1.72.0 @xmath27 , and minor features throughout the h - band are likely attributable to weaker h@xmath0o lines .
two weak features at 1.573 and 1.626 @xmath27 are present , possibly attributable to feh absorption @xcite or poor correction of telluric oh lines ; these features are not coincident with the 1.6 @xmath27 ch@xmath30 band .
overall , this portion of the spectrum of 2mass 0532 + 8246 is quite unlike any known late - type m , l , or t dwarf , based on its blue slope and absence of ch@xmath30 and co bands .
the j - band spectrum , shown in detail in figure 2b , is far more rich , with atomic lines of k i ( 1.1690 , 1.1773 , 1.2543 , 1.2522 @xmath27 ) and fe i
( 1.1528 , 1.1886 , 1.1976 , 1.2128 , 1.3210 @xmath27 ) , and strong molecular bands of feh ( 1.194 , 1.239 @xmath27 ) and h@xmath0o ( 1.3 @xmath27 ) observed . above the spectrum we plot an opacity spectrum of feh @xcite , which shows that much of the fine structure observed is caused by this molecule .
there are no lines of mg i , ca i , or al i present in this spectral region , and the @xmath31 tio system around 1.25 @xmath27 @xcite is either absent or obscured by feh absorption .
the overall spectral morphology at j - band is quite similar to the l7 denis 0205@xmath21159ab , with the notable exception of stronger feh and h@xmath0o absorption .
we obtained a red optical spectrum of 2mass 0532 + 8246 on 3 january 2003 ( ut ) using lris on the keck - i 10 m telescope .
conditions were clear with sub - arcsecond seeing . two exposures of 1200 sec each were obtained through the red channel , dithered by 2@xmath29 between exposures along the 1@xmath29 ( 4.7 pixels ) slit .
we employed the 400 lines mm@xmath6 grating blazed at 8500 , yielding 630010100 spectra with 7 resolution ( r @xmath32 1200 ) .
dispersion on the chip was 1.9 pixel@xmath6 .
the og570 order - blocking filter was used to suppress higher - order light .
observations of the b1 v flux standard hiltner 600 @xcite were obtained for flux calibration , and the g0 v star hd 38847 was observed after the target at similar airmass ( 2.4 ) for telluric calibration .
henear arc lamp exposures were taken immediately after the target observations for wavelength calibration , and quartz lamp flat - field exposures ( reflected off of the interior dome ) were observed at the start of the night to correct for detector response .
data reduction procedures , using standard iraf routines , are discussed in detail in @xcite .
the final flux - calibrated and telluric - corrected optical spectrum for 2mass 0532 + 8246 is included in figure 2a and shown in detail in figure 2c .
strong metal hydride bands of feh ( 8692 , 9896 ) , crh ( 8611 , 9969 ) , and cah ( 6750 ) are clearly discernable .
the 9896 feh wing - ford band is the strongest seen in any cool spectrum to date , and absorption from the shorter wavelength feh and crh bands extends to roughly 9100 .
weaker bands of tio ( 7053 , 8432 ) and h@xmath0o ( 9250 , 9400 ) are also seen .
we point out an absorption feature between 95709700 which is too strong to be attributable to intrinsic h@xmath0o absorption , based on the weakness of the 9250 band ; it is also not properly placed for feh , crh , tih , or ch@xmath30 absorption .
this band is weakly present in late - type m and early - type l dwarf spectra , and may be a hitherto unrecognized metal hydride band .
it is also possible , although unlikely , that this feature is attributable to residual telluric absorption , as telluric corrections were made only to 9650 ( the extent of the strong terrestrial h@xmath0o bands ; stevenson 1994 ) .
individual alkali lines of na i ( 8183 , 8195 ) , rb i ( 7800 , 7948 ) , and cs
i ( 8521 , 8943 ) are present , while the resonance k i doublet ( 7665 , 7699 ) is exceptionally strong and pressure - broadened , as seen in late - type l and t dwarfs @xcite .
the red optical / nir colors of 2mass 0532 + 8246 are clearly a consequence of the k i red wing ; integrating the spectrum over the @xmath33 bandpass @xcite yields @xmath33 @xmath28 19.2 and @xmath34 @xmath28 4.0 , similar to mid- and late - type l dwarfs @xcite .
the blue end of the spectrum appears to be similarly suppressed by the pressure - broadened na i doublet ( 5890 , 5896 ) .
both li i ( 6708 ) absorption and h@xmath35 ( 6563 ) emission lines are absent .
the overall spectral morphology , excluding hydride bands and tio , is again quite similar to the l7 denis 0205@xmath21159ab , suggesting a rather cool atmosphere .
2mass 0532 + 8246 is not a t dwarf , based on the absence of the 1.6 and 2.2 @xmath27 ch@xmath30 bands @xcite , and the presence of strong feh and crh and weak tio and na i features in the optical @xcite .
it appears that this object is instead a metal - deficient , late - type l dwarf .
enhanced cia h@xmath0 and strong hydride bands in the optical and nir spectra of 2mass 0532 + 8246 are consistent with a metal - poor atmosphere .
both are consequences of the overall reduced metal opacity , resulting in larger column abundances of the remaining chemical species and increased relative abundance of metal hydrides over double - metal species .
the presence of the 7053 and 8432 tio bands is somewhat at odds with this interpretation , particularly since tio is generally weak or absent in disk dwarfs later than l5 @xcite .
one possibility is that the mechanism of ti depletion in these objects incorporation into multiple - metal condensates of catio@xmath36 , ca@xmath36ti@xmath0o@xmath37 , ca@xmath30ti@xmath36o@xmath38 , and ti@xmath0o@xmath36 @xcite may be inhibited in the low - metallicity atmosphere of 2mass 0532 + 8246 .
chemical equilibrium calculations incorporating subsolar abundances are required to address this hypothesis .
the tio bands may also indicate a somewhat warmer atmosphere than the l7 comparison object denis 0205@xmath21159ab . classifying 2mass 0532
+ 8246 is therefore not straightforward , particularly as there are no other metal - deficient l dwarfs currently known for direct comparison .
the m subdwarf and extreme subdwarf sequences of @xcite terminate at sdm7 and esdm5.5 , respectively , although recent late - type discoveries are classified via extrapolation and limited modifications of this scheme @xcite .
metal classification ( d , sd , esd ) in the gizis scheme is based on the strength of the 7053 tio band , which is exceedingly weak in the spectrum of 2mass 0523 + 8246 .
numerical classification is based on the 6750 cah band , which is suppressed in this spectrum by na i and k i absorption .
thus , a simple extrapolation of the gizis scheme fails in the l dwarf regime . while there is general agreement between the spectral morphologies of 2mass 0532 + 8246 and the denis 0205@xmath21159ab
, the optical / j - band hydride features and tio bands are more similar to mid - type l dwarfs @xcite .
additionally , there are no cool dwarf standards that match the 1.32.4 @xmath27 spectral region .
we therefore tentatively classify this object a late - type sdl .
identification of additional metal - poor l subdwarfs may eventually allow the definition of a robust classification scheme .
examination of the ircam and ccd followup images show that 2mass 0532 + 8246 has moved significantly since it was first imaged by 2mass .
we therefore measured its proper motion by deriving its coordinates on each of the images , using the 2mass coordinates of background stars for astrometric calibration .
results are listed in table 2 .
a linear fit yields @xmath3 = 2@xmath460@xmath50@xmath415 yr@xmath6 at position angle [email protected]@xmath21 , making 2mass 0532 + 8246 one of the highest proper motion stars known @xcite .
this object is not present in the lhs because of its optical faintness , although a faint counterpart may be present in the poss - ii r - band image at the extrapolated position ( figure 1 ) .
residuals in the proper motion fit are indicative of parallactic motion , but positional uncertainties are too large for a reliable measure . the radial velocity of 2mass 0532 + 8246 was derived by cross - correlating its j - band spectrum with those of five early- and mid - type l dwarfs ( chosen for their strong nir feh absorption ) with measured radial velocities : 2mass 0746 + 2000ab , 2mass 1439 + 1929 , g 196 - 3b , kelu 1 , and denis 1228@xmath21547ab @xcite .
all five of these objects have been observed as part of the nirspec brown dwarf spectroscopic survey @xcite , and are therefore wavelength calibrated in the same manner as the data presented here .
we derive a heliocentric @xmath7 km s@xmath6 , where the uncertainty is derived from the scatter of values amongst the comparison objects .
this value is consistent with heliocentric velocity offsets in the 1.1690/1.1773 @xmath27 k i lines ( @xmath39 km s@xmath6 ) and the optical cs i and rb i lines ( @xmath40 km s@xmath6 ) .
the substantial radial velocity and proper motion of 2mass 0532 + 8246 identifies it as a high velocity source .
determining its total space motion , however , requires an estimate of its distance .
if we assume 2mass 0532 + 8246 is bound to the galaxy ( galactocentric @xmath41 km s@xmath6 ; carney , latham , & laird 1988 ) , we derive an upper limit distance of 54 pc assuming ( @xmath42)@xmath9 = ( 10.00 , 5.25 , 7.17 ) km s@xmath6 @xcite and local standard of rest ( lsr ) velocity @xmath43 km s@xmath6 @xcite .
a spectrophotometric parallax estimate is highly uncertain , given the substantial redistribution of flux in the nir and uncertainty in the numeric subtype for this object .
however , assuming that the @xmath44 magnitude of 2mass 0532 + 8246 is similar to that of mid- to late - type l dwarfs ( * ? ? ?
* @xmath45 ) , we derive a conservative estimate of 1030 pc
. the corresponding heliocentric space velocities at the median distance are ( @xmath42 ) = ( @xmath213 , @xmath2301 , 28 ) km s@xmath6 .
the substantial v velocity , between @xmath2390 and @xmath2212 km s@xmath6 over the adopted distance range , indicates zero or retrograde motion with respect to the lsr , and lends additional support that this object is a member of the halo population .
disk l dwarfs with spectral types later than @xmath32l4 are predominately substellar @xcite ; hence , the late - type spectral morphology of 2mass 0532 + 8246 suggests that it too may be a brown dwarf . to test this hypothesis , we compared the estimated t@xmath46 of this object to subsolar metallicity evolutionary models from @xcite . in figure 3 , we plot t@xmath46 versus mass for ages 10 and 15 gyr and metallicities @xmath47 = 0.1 , 0.3 , and 0.01 @xmath11 . at these late ages ,
the transition between stellar and substellar masses spans a substantial range in temperature , and is more dramatic for lower metallicities due to the reduced atmospheric opacities and hence enhanced luminosities @xcite . deriving an accurate t@xmath46 for 2mass
0532 + 8246 is difficult without adequate distance or bolometric flux measurements .
solar - metallicity mid- to late - type l dwarfs have temperatures @xmath48 k @xcite . in general
there is little difference between the temperatures of m - type subdwarfs and dwarf stars of the same numeric type @xcite , although late - type m subdwarfs ( e.g. , the sdm7 lhs 377 ) may be 100 - 300 k hotter @xcite .
we therefore adopt a conservative temperature range of @xmath49 k. we further assume that the metallicity of 2mass 0532 + 8246 lies in the range @xmath50 @xmath11 , typical for sdm and esdm stars @xcite . for these parameters , we derive a mass range of 0.077 @xmath8 m @xmath8 0.085 m@xmath9 .
larger masses correspond to lower metallicities , so that in each case the derived mass is below the hbmm .
therefore , we conclude the 2mass 0532 + 8246 is probably substellar , although more refined estimates of its atmospheric properties through further observations and spectral modelling are required to validate this result .
the existence of a substellar subdwarf is not unexpected , as mass functions for the nearby halo population @xcite and globular clusters @xcite are shallow but rising : dn / dm @xmath51 m@xmath52 at m @xmath32 0.15 m@xmath9 , similar to the disk population @xcite .
however , the steep drop in t@xmath46 , and hence luminosity , across the metal - poor substellar boundary implies that the vast majority of these halo brown dwarfs will have t@xmath53 k ; i.e. , in the t dwarf regime and cooler @xcite .
indeed , one possible t - type subdwarf has already been identified , 2mass 0937 + 2931 @xcite , an object which also exhibits unusually strong cia h@xmath0 and feh absorption @xcite and a substantial proper motion @xcite .
relatively few l - type subdwarfs should exist , however , as they span a much narrower range in mass at ages later than 10 gyr .
the discovery of 2mass 0532 + 8246 may therefore be quite fortuitous .
given its optical faintness , we expect that few counterparts to this late - type l subdwarf will be identified in current proper motion surveys .
nir proper motion surveys , on the other hand , could potentially reveal more of these objects by detecting them at the peak of their spectral energy distribution .
a sizeable fraction of the sky ( @xmath3230% ) has been scanned more than once by 2mass during its three years of observations , and analysis of these data may identify new , very cool high - motion stars . a second generation
2mass survey could greatly improve on this work , adding substantially to the short list of very cool subdwarfs currently known .
a. j. b. acknowledges useful discussions with k. cruz and j. wei - chun during the preparation of the manuscript , and thanks our anonymous referee for her / his helpful criticisms .
we are grateful for the assistance of our keck observing assistants joel aycock and gary puniwai , and instrument specialist paola amico ; and our palomar telescope operators , karl dunscombe and barrett `` skip '' staples . we also acknowledge the 2mass staff for their laudable efforts on the 2mass database , without which none of this research could have been possible .
a. j. b. acknowledges support by nasa through hubble fellowship grant hst - hf-01137.01 awarded by the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, for nasa , under contract nas 5 - 26555 .
a. s. b. acknowledges funding through nasa grants nag5 - 10760 and nag5 - 10629 .
portions of the data presented herein were obtained at the w. m. keck observatory which is operated as a scientific partnership among the california institute of technology , the university of california , and the national aeronautics and space administration .
the observatory was made possible by the generous financial support of the w. m. keck foundation .
this publication makes use of data from the two micron all sky survey , which is a joint project of the university of massachusetts and the infrared processing and analysis center , funded by the national aeronautics and space administration and the national science foundation .
2mass data were obtain through the nasa / ipac infrared science archive , which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
poss - i and poss - ii images were obtained from the dss image server maintained by the canadian astronomy data centre , which is operated by the herzberg institute of astrophysics , national research council of canada .
the authors wish to extend special thanks to those of hawaiian ancestry on whose sacred mountain we are privileged to be guests .
lc @xmath55 & 5@xmath56 32@xmath57 53@xmath446 + @xmath58 & + 82@xmath21 46@xmath57 46@xmath45 + @xmath59 & 19.2 + gunn @xmath20 & [email protected] + 2mass @xmath60 & [email protected] + 2mass @xmath61 & [email protected] + 2mass @xmath62 & [email protected] + @xmath3 & 2@xmath460@xmath50@xmath415 yr@xmath6 + @xmath63 & [email protected]@xmath21 + @xmath64 & @xmath65 km s@xmath6 + llcccc 01 mar 1999 02:20 & 2mass & 5@xmath56 32@xmath57 53@xmath446 & + 82@xmath21 46@xmath57 46@xmath453 & 0@xmath430 & 0@xmath430 + 25 sep 1999 11:53 & p60 ircam & 5@xmath56 32@xmath57 54@xmath414 & + 82@xmath21 46@xmath57 45@xmath460 & 0@xmath407 & 0@xmath406 + 18 nov 2002 08:01 & p60 ccd & 5@xmath56 32@xmath57 57@xmath437 & + 82@xmath21 46@xmath57 40@xmath433 & 0@xmath435 & 0@xmath423 + 25 jan 2003 06:33 & p60 ccd & 5@xmath56 32@xmath57 57@xmath466 & + 82@xmath21 46@xmath57 40@xmath400 & 0@xmath437 & 0@xmath422 + | we present the discovery of the first l - type subdwarf , 2mass j05325346 + 8246465 .
this object exhibits enhanced collision - induced h@xmath0 absorption , resulting in blue nir colors ( @xmath1 ) .
in addition , strong hydride bands in the red optical and nir , weak tio absorption , and an optical / j - band spectral morphology similar to the l7 denis 0205@xmath21159ab imply a cool , metal - deficient atmosphere .
we find that 2mass 0532 + 8246 has both a high proper motion , @xmath3 = 2@xmath460@xmath50@xmath415 yr@xmath6 , and a substantial radial velocity , @xmath7 km s@xmath6 , and its probable proximity to the sun ( d = 1030 pc ) is consistent with halo membership . comparison to subsolar - metallicity evolutionary models strongly suggests that 2mass 0532 + 8246 is substellar , with a mass of 0.077 @xmath8 m @xmath8 0.085 m@xmath9 for ages 1015 gyr and metallicities @xmath10 @xmath11 .
the discovery of this object clearly indicates that star formation occurred below the hydrogen burning mass limit at early times , consistent with prior results indicating a flat or slightly rising mass function for the lowest - mass stellar subdwarfs .
furthermore , 2mass 0532 + 8246 serves as a prototype for a new spectral class of subdwarfs , additional examples of which could be found in nir proper motion surveys . |
hyperspectral images consist of very high - dimensional pixel observations that allow reconstruction of the spectral profiles of objects imaged thanks to the acquisition of several hundred narrow spectral bands .
the supervised classification of these pixels is a challenging task , which commonly arises in remote sensing imaging @xcite .
structure of the hyperspectral imagery is seldom studied in a comprehensive manner , with most approaches focusing either on spatial information building on tools available for normal imagery or with a focus on spectral information without a principled way to make use of both .
we propose a novel approach to classification based on kernel embeddings of distributions which utilizes both the spatial and spectral information in the data . while aimed at hyperspectral imaging , the method we propose is general and can be applied to other types of data .
kernel methods and support vector machines have been employed in the hyperspectral imaging in @xcite the pixel data is lifted into a potentially infinite - dimensional feature space , called reproducing kernel hilbert space ( rkhs ) , where linear separating hyperplanes are sought .
however , spectral information contained in pixels is often not sufficient for such task , and as we will see , including the local / spatial information available in the imagery is key to obtain good classification accuracy .
our approach is to encode the spatial neighbourhood of each pixel as a random sample from a distribution associated to that pixel and to treat such distribution as an additional feature for classification . in order to add consistent spatial information
, we also use the hadamard multiplication of two kernels .
where one kernel is the kernel embeddings of distributions , and the other one is the linear kernel of spatial information similarly to @xcite . in section [ sec : related_work ] , related work is reviewed .
section [ sec : background ] provides the background on kernel embeddings of distributions , random features for fast approximations to kernel methods , and on mathematical morphology , which allow us to analyse and understand the geometrical structures of images .
section [ sec : theory ] studies the consistency and convergence rate of the proposed method and experiments are given in section [ sec : experiments ] .
many techniques aim to include the spatial information in the classification process . of particular interest
are those combining feature space representations describing the spatial information with those describing the pixels .
morphological feature spaces have been considered in several publications , with impressive results @xcite . on the other hand ,
kernel methods have also been studied extensively , and more particularly the compositions of kernels @xcite , which allow building new feature space representations .
we marry these approaches with a framework of @xcite , where instead of the usual feature map , sending each data point to the feature space , a whole distribution can be represented in the rkhs .
this yields a framework for learning on distributions via their representations in this rkhs . in our approach
, each pixel is associated to a distribution of its neighbours effectively , a hyperspectral image is treated as a set of such distributions .
this is similar to the approach to regression applied in @xcite to the multispectral imaging data .
however , the authors of @xcite partition a multispectral image and classify the partitions - with a goal to obtain responses at the level of the _ groups of neighbouring pixels _ , which suffices when the goal is to predict an averaged quantity of an image area ( e.g. aerosol concentration as studied in @xcite ) and the pixel - level classification is not considered .
another related line of work is that of @xcite , where they used the mean map on hyperspectral to perform a dimensionality reduction .
let @xmath0 be a positive definite kernel . by moore - aronszajn theorem @xcite , there is a unique rkhs @xmath1 of real - valued functions on @xmath2 where @xmath3 , for all @xmath4 , implying that @xmath5 corresponds to an inner product between features and , in particular , @xmath6 .
this means that @xmath7 can be viewed as a feature of @xmath8 . for many typical choices of kernels @xmath5 ,
the rkhs @xmath1 is infinite - dimensional .
now , let @xmath9 denote a random variable following a distribution @xmath10 .
the _ mean map _ or the _ kernel embedding _
@xcite of @xmath10 is defined as : @xmath11 = \int_{{\mathcal x}}k(\cdot , x ) \
\mathrm{d}\mathcal{p}(x ) , \end{aligned}\ ] ] where the expectation is over @xmath1 . for _ characteristic _ kernels @xcite , which include gaussian rbf , matern family and many others
, this embedding is injective on the space of all probability distributions ( i.e. captures information on all moments , akin to a characteristic function ) .
further , if we are given two random variables , @xmath9 following the distribution @xmath10 , and @xmath12 following the distribution @xmath13 , the inner product between the corresponding embeddings is given as @xmath14,\end{aligned}\ ] ] which is sometimes referred to as a _
mean map kernel_. for a random sample @xmath15 , drawn independently and identically distributed from @xmath10 , we can define the empirical mean map : @xmath16 and for random samples @xmath15 from @xmath10 and @xmath17 from @xmath13 , we obtain the empirical mean map kernel : @xmath18 the computational and storage requirements for kernel methods on large datasets can be prohibitive in practice due to the need to compute and store the kernel matrix . if we consider a dataset of @xmath19 @xmath20-dimensional observations , the storage requirements are @xmath21 and the calculation takes @xmath22 operations .
a remedy developed by @xcite is to approximate translation - invariant kernels in an unbiased way using a random feature representation .
namely , any translation - invariant positive definite kernel @xmath5 , such that @xmath23 , @xmath24 can be written as @xmath25 @xmath26 $ ] , where @xmath27 follows some distribution @xmath28 ( spectral measure of the kernel ) .
thus , by sampling i.i.d .
vectors @xmath29 from @xmath28 , we can approximate kernel @xmath5 by @xmath30 defined by : @xmath31 @xmath32 , so that the original feature map @xmath33 , potentially living in an infinite - dimensional space , is approximated by an explicit @xmath34dimensional feature vector : @xmath35 thus , the mean map and the mean map kernel can be estimated using these finite - dimensional representations . in this contribution , we will focus on gaussian rbf kernels for which the spectral measure @xmath28 is also gaussian .
let us now turn our attention to a hyperspectral image @xmath36 . around each pixel location @xmath37 , we consider a square patch @xmath38 of size @xmath39 where we will treat the pixels as a random sample from a distribution @xmath40 specific to the location @xmath37 . instead of calculating the kernel between individual data points , we will calculate kernel between these distributions .
an empirical mean map kernel is thus given simply by : @xmath41 where @xmath42 denotes the measurement vector at location @xmath43 and in the last line we employ a random feature approximation of @xmath5 .
it should be noted that there may be outliers in a patch , which can damage the estimation of the mean .
similarly to the work of @xcite , we proposed to use a weighted mean map , where the weights depend on spatial information .
the kernels we obtain , called convolutional kernels , have also been used in @xcite .
in contrast to @xcite , however , we will use random feature expansions to explicitly represent the feature space .
the convolutional kernel is defined as : @xmath44 where @xmath45 represents a normalised version of @xmath36 , such that for all @xmath46 $ ] = @xmath47 .
so we do a product of a kernel on the positions , another on the magnitudes , and a third one is an rbf kernel between spectra .
this formula can be interpreted as a weighted mean map which is defined by : @xmath48 where @xmath49 is a positive definite kernel arising from the random feature space expansion .
let us consider that the data are partitioned into sets following the same distribution , then the structure of our data is given by @xmath50 with @xmath51 , where @xmath52 are drawn from a joint meta distribution @xmath53 .
we follow the notation of @xcite .
let us denote @xmath54 a loss function .
let us write the following expected risk function of the data for the svm problem : @xmath55 we can modify it to mean map embedding classification problem : @xmath56 we can also write the empirical risk function , for mean map embedding classification problem : @xmath57 finally we can also write the empirical risk function , for the empirical mean map embedding classification problem : @xmath58 then we would like to obtain an inequality between @xmath59 and @xmath60 . to do that , inspired by @xcite
, we derive a inequality @xmath59 and @xmath61 : + given that @xmath62 an arbitrary probability distribution with variance @xmath63 , a lipschitz continuous function @xmath64 with constant @xmath65 , an arbitrary loss function @xmath66 that is lipschitz continuous in the second argument with constant @xmath67 , it follows that : @xmath68 the proof of this theorem can be found on the supplementary materials
. then we might use @xcite where we have an inequality between @xmath59 and @xmath69 : let @xmath70 denote the loss class , let @xmath71 denote the rademacher complexity .
let @xmath72 be a bound on the variance of the functions in @xmath73 .
if the trace of the kernel is bounded , the loss function @xmath74 that is lipschitz continuous , for any @xmath75 , the following bound holds with probability at least @xmath76 @xmath77 given that @xmath64 is a lipschitz continuous function with constant @xmath65 , an arbitrary loss function @xmath66 that is lipschitz continuous with constant @xmath78 , it follows that : @xmath79 the proof of this theorem can be found on the supplementary materials .
we also need the following theorem proved in @xcite assume that @xmath80 for all @xmath81 with @xmath82 , and that @xmath5 is an universal kernel . then with probability at least @xmath83 : @xmath84 - \mu[x]| \leq 2\mathcal{r}_n(\mathcal{h } , p ) + r\sqrt{\log\left(1/\delta\right)/n}\end{aligned}\ ] ] where @xmath85 denotes the rademacher average associated with @xmath86 and @xmath1 . then by combining the previous theorems we easily have the following theorem .
[ theorem1 ] given the conditions of the previous theorems .
then with probability at least @xmath83 : @xmath87 \\ + 8 \mathcal{r}_n(\mathcal{g } ) + \sigma(\mathcal{g } ) \sqrt{\frac{8 \log(2/\delta)}{n}}+ \frac{3 \log(2/\delta)}{n}\end{aligned}\ ] ]
we evaluate the classification accuracy of the proposed approach using two standard datasets : the aviris indian pines , and the rosis university of pavia .
the first data set is an image of dimension @xmath88 pixels , with @xmath89 spectral bands and its geometrical resolution is of 3.7 m. the training set is composed of @xmath90 pixels , and the image is composed of @xmath91 classes .
the dimensions of the second data set are @xmath92 pixels , with @xmath93 spectral bands and its geometrical resolution is of 1.3 m. the training set is composed of @xmath94 pixels and a testing set of @xmath95 pixels , and the image is composed of @xmath96 classes .
there is a commonly used testing set for pavia data , and we report performance on this testing set . in the first data set
there is no testing set so we generate 20 monte - carlo simulations , selecting randomly 5 pixels per class , then aggregate the result of the classification .
we used the morphological profile ( mp ) feature @xcite space which is commonly used in pixel classifiation and is described in the supplementary material .
we also use the product of the two kernels where one is the mp kernel and the other is the kmm kernels .
this kind of techinique has been previously explored in @xcite .
in contrast to the previous work , which approximate this product of kernels thanks to addition of kernels , we can do the real multiplication since we work with finite dimension hilbert spaces .
the results of classification are reported in table [ table1 ] ( indian pine ) , and table [ table2 ] ( pavia ) .
the classification algorithm used is the c - svm @xcite were the parameter @xmath97 was selected with @xmath98fold cross - validation on a grid @xmath99,with @xmath100 $ ] .
the results on table [ table1 ] and [ table2 ] show us that kernel mean map can perform as well as state of art results on these images .
the size of scale @xmath39 seems to be important , this is linked with the theorem [ theorem1 ] , where we see that increasing the size of the scale increase in a way the size of the training set .
an explanation of the parameters of evaluation can be found in @xcite ( pages 166 - 167 ) and also in the supplementary materials .
.overall accuracy , kappa statistic , and average accuracy obtained for different kernels , applied on the aviris indian pines hyperspectral data set .
i have run 20 monte carlo simulations .
i have selected on the training set just 5 samples per class . [ cols= " < , < , < , < , < " , ]
in this article we developed a new method for pixel classification in hyperspectral imaging .
the method uses spatial information encoded through distributions of the neighbourhood around each pixel .
even with very simple kernel choices ( gaussian rbf applied directly to raw data ) the obtained results are comparable with state - of - the - art .
we establish convergence rates for , and prove that we are two - stage consistent .
further improvements are possible by using different feature spaces employing suitable representations of the individual pixels . we believe that we have established an interesting research direction where local distributions are treated as additional features for a supervised learning task , which is of particular interest in hyperspectral imaging where it is difficult to combine spatial and spectral information in a principled way and our approach can be viewed as a step in that direction .
10 gustavo camps - valls , devis tuia , lorenzo bruzzone , and jon atli benediktsson , `` advances in hyperspectral image classification : earth monitoring with statistical learning methods , '' , vol .
1 , pp . 4554 , 2014 . j anthony gualtieri and robert f cromp , `` support vector machines for hyperspectral remote sensing classification , '' in _ the 27th aipr workshop : advances in computer - assisted recognition_. international society for optics and photonics , 1999 , pp
. 221232 .
m. fauvel , j.a .
benediktsson , j. chanussot , and j.r .
sveinsson , `` spectral and spatial classification of hyperspectral data using svms and morphological profiles , '' , vol .
46 , no . 11 , pp . 38043814 , nov 2008 .
mathieu fauvel , yuliya tarabalka , jon atli benediktsson , jocelyn chanussot , and james c tilton , `` advances in spectral - spatial classification of hyperspectral images , '' , vol .
3 , pp . 652675 , 2013 .
leyuan fang , shutao li , wuhui duan , jinchang ren , and jn atli benediktsson , `` classification of hyperspectral images by exploiting spectral spatial information of superpixel via multiple kernels , '' , vol .
12 , pp . 66636674 , 2015 .
jun li , prashanth reddy marpu , antonio plaza , jos m bioucas - dias , and jon atli benediktsson , `` generalized composite kernel framework for hyperspectral image classification , '' , vol . 51 , no .
9 , pp . 48164829 , 2013 .
gustavo camps - valls , luis gomez - chova , jordi muoz - mar , joan vila - francs , and javier calpe - maravilla , `` composite kernels for hyperspectral image classification , '' , vol .
1 , pp . 9397 , 2006 .
mathieu fauvel , , ph.d .
thesis , institut national polytechnique de grenoble - inpg ; universit dislande , 2007 .
m. dalla mura , j.a .
benediktsson , b. waske , and l. bruzzone , `` morphological attribute profiles for the analysis of very high resolution images , '' , vol .
10 , pp . 37473762 , oct 2010 .
martino pesaresi and j.a .
benediktsson , `` a new approach for the morphological segmentation of high - resolution satellite imagery , '' , vol .
309320 , feb 2001 .
sebastien lefevre , laetitia chapel , and francois merciol , `` hyperspectral image classification from multiscale description with constrained connectivity and metric learning , '' june 2014 , pp .
jun li , xin huang , paolo gamba , jos m bioucas - dias , liangpei zhang , jon atli benediktsson , and antonio plaza , `` multiple feature learning for hyperspectral image classification , '' , vol .
3 , pp . 15921606 , 2015 .
fauvel mathieu , chanussot jocelyn , benediktsson jn atli , et al .
, `` kernel principal component analysis for the classification of hyperspectral remote sensing data over urban areas , '' , vol .
2009 , 2009 .
a. smola , a. gretton , l. song , and b. schlkopf , `` a hilbert space embedding for distributions , '' in _ alt _ , 2007 ,
krikamol muandet , kenji fukumizu , francesco dinuzzo , and bernhard schlkopf , `` learning from distributions via support measure machines , '' in _ advances in neural information processing systems _ , 2012 , pp .
zoltn szab , arthur gretton , barnabs pczos , and bharath sriperumbudur , `` two - stage sampled learning theory on distributions , '' , 2014 .
junier b oliva , dougal j sutherland , barnabs pczos , and jeff schneider , `` deep mean maps , '' , 2015 .
michele volpi and devis tuia , `` spatially aware supervised nonlinear dimensionality reduction for hyperspectral data , '' june 2014 , pp .
a. berlinet and c. thomas - agnan , , kluwer , 2004 .
b. sriperumbudur , a. gretton , k. fukumizu , g. lanckriet , and b. schlkopf , `` hilbert space embeddings and metrics on probability measures , '' , vol . 11 , pp . 15171561 , 2010 .
b. sriperumbudur , k. fukumizu , and g. lanckriet , `` universality , characteristic kernels and rkhs embedding of measures , '' , vol .
12 , pp . 23892410 , 2011 .
a. rahimi and b. recht , `` random features for large - scale kernel machines , '' in _ advances in neural information processing systems _ , 2007 , pp
. 11771184 .
seth r flaxman , yu - xiang wang , and alexander j smola , `` who supported obama in 2012 ?
: ecological inference through distribution regression , '' in _ proceedings of the 21th acm sigkdd international conference on knowledge discovery and data mining_. acm , 2015 , pp .
289298 .
julien mairal , piotr koniusz , zaid harchaoui , and cordelia schmid , `` convolutional kernel networks , '' in _ advances in neural information processing systems _ , 2014 , pp .
26272635 .
corinna cortes , marius kloft , and mehryar mohri , `` learning kernels using local rademacher complexity , '' in _ advances in neural information processing systems _ , 2013 , pp .
27602768 .
chih - chung chang and chih - jen lin , `` libsvm : a library for support vector machines , '' , vol .
27:127:27 , 2011 , software available at http://www.csie.ntu.edu.tw/ cjlin / libsvm .
jean serra , , academic press , inc . ,
orlando , fl , usa , 1983 .
mathematical morphology operators are non - linear image processes based on the spatial structure of the image .
let @xmath101 be a grey scale image which can be represented by a function .
two basic operators in morphology are the grey - level erosion and the grey - level dilatation whose definition are respectively given by @xcite : @xmath102 where @xmath103 is a structuring function , which introduces the effect of the operators by the geometry of its support as well as the penalizations .
we consider for simplicity uniform structuring functions which are formalised by their support set or shape @xmath104 , called structuring element . by concatenation of these two basic morphological operators
it is possible to obtain more evolved operators such as the opening and the closing @xcite : these operators remove from @xmath101 all the bright ( opening ) or dark ( closing ) structures where the structuring element @xmath104 can not fit .
however they also modify the value of pixels when @xmath104 fit .
thus to avoid these artefacts it has been proposed in @xcite to use geodesic opening and closing . then by considering a set @xmath106 , @xmath107 , of indexed geodesic openings , and a set @xmath108 , @xmath107 , of indexed geodesic closings where typically , the index @xmath109 is associated to the size of the structuring element .
then thanks to the granulometry axiomatic @xcite we obtain a scale space representation of an image , which allows an image structures decomposition . then the moprhological profile ( mp ) of a grey scale image @xmath101
is defined as @xmath110 dimension vector : to be able to use the mp on hyperspectral images we first reduce the dimension of the data thanks to pca , and then project the data on a @xmath112 dimensional space which is of smaller dimension than the original space . so a hyperspectral image is represented by @xmath112 grey scale images , then on each of these images we calculate the mp and we concatenate them . hence , the spatial feature space is of dimension @xmath113 .
the formula for the convolutional kernel mean map is defined in equation ( [ kcn1 ] ) by : @xmath114 this equation can be rewritten as : @xmath115 where @xmath116 corresponds to the position of the pixel , whereas @xmath117 is a pixel value and so a spectrum of dimension @xmath20 . for the following formula
we write @xmath118 and @xmath119 respectively the first and the second spatial coordinate , and @xmath120 the @xmath109-th coordinate of the vector @xmath42 .
finally @xmath121 is defined by : @xmath122 then we use the random feature trick on this new vector . given that @xmath62 an arbitrary probability distribution with variance @xmath63 , a lipschitz continuous function @xmath64 with constant @xmath65 , an arbitrary loss function @xmath66 that is lipschitz continuous in the second argument with constant @xmath67 , it follows that : @xmath68 @xmath123 $ ] + @xmath124
since @xmath54 is lipschitz continuous we have : @xmath125 .
+ thanks to the cauchy - schwarz inequality we have : + @xmath126 since @xmath101 is lipschitz continuous we have : + @xmath127 @xmath129 $ ] + @xmath130 since @xmath54 is lipschitz continuous we have : @xmath131 . +
thanks to the cauchy - schwarz inequality we have : + @xmath132 since @xmath101 is lipschitz continuous we have : + @xmath133 [ [ assessing - the - accuracy - of - the - classification - on - hyperspectral - data ] ] assessing the accuracy of the classification on hyperspectral data ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ most of the time , when one does a classification on hyperspectral data , we do not face a binary classification .
hence we have to handle a more tricky classification .
to evaluate this classification we use what is called a confusion matrix .
each column of the matrix represents the number of occurrences of an estimated class , while each row represents the number of occurrences of a real class .
if we write @xmath97 the confusion matrix of a classification then @xmath134 is the number of pixels of class @xmath109 assign to the class @xmath135 by the classifier .
let us write @xmath136 the number of class . | we propose a novel approach for pixel classification in hyperspectral images , leveraging on both the spatial and spectral information in the data .
the introduced method relies on a recently proposed framework for learning on distributions by representing them with mean elements in reproducing kernel hilbert spaces ( rkhs ) and formulating a classification algorithm therein .
in particular , we associate each pixel to an empirical distribution of its neighbouring pixels , a judicious representation of which in an rkhs , in conjunction with the spectral information contained in the pixel itself , give a new explicit set of features that can be fed into a suite of standard classification techniques
we opt for a well established framework of support vector machines ( svm ) .
furthermore , the computational complexity is reduced via random fourier features formalism .
we study the consistency and the convergence rates of the proposed method and the experiments demonstrate strong performance on hyperspectral data with gains in comparison to the state - of - the - art results .
+ * keywords : * hyperspectral images , pixelwise classification , kernel methods .
[ theorem]theorem 3 [ theorem]theorem 1 [ theorem]definition [ theorem]proposition [ 1][proof]*#1 . * ' '' '' |
hh 34 is one of the hh objects in herbig s catalogue ( herbig 1974 ) .
it jumped into the limelight with the paper of reipurth et al .
( 1986 ) , who showed that the hh object ( hh 34s ) had a bow shaped morphology , and a jet - like association of aligned knots ( pointing towards the apex of the bow shock ) .
later observations showed the existence of a northern counterpart to hh 34s ( hh34n ) and of a series of bipolar bow shock pairs at larger distances from the outflow source ( bally & devine 1994 ; eislffel & mundt 1997 ; devine et al .
there is a wealth of observations of this outflow , including images , spectrophotometry , radial velocity and proper motion measurements at optical ( see , e. g. , heathcote & reipurth 1992 ; eislffel & mundt 1992 ; morse et al .
1992 , 1993 ; reipurth et al . 2002 ; beck et al . 2007 ) and ir wavelemgths ( stapelfeldt et al . 1991
; stanke et al . 1998 ; reipurth et al .
these observations show that the hh 34 outflow ( at a distance of @xmath3 pc , see menten et al .
2007 ) has a plane of the sky velocity of @xmath4 km s@xmath5 and propagates at an angle of @xmath6 from the plane of the sky ( the southern lobe being directed towards the observer ) .
the flow itself is driven by hh 34 irs , a class i protostar surrounded by a relatively large ( @xmath7 au radius ) circumstellar disk ( stapelfeldt & scoville 1993 ; anglada et al .
the possible coupling of the disk and outflow has motivated the search for signatures of internal rotation within the jet as consequence of the transfer of angular momentum from the rotating disk into the highly collimated jet ( coffey et al .
2010 ) . in spite of the clear bipolar symmetry of the bow shock pairs ( extending to @xmath8 pc from the source ,
see devine et al .
1997 ) , observations extending over @xmath9 decades did not detect a nothern counterpart for the chain of aligned knots extending @xmath10 southwards from the hh 34 source .
this situation changed with the paper of garca lpez et al .
( 2008 ) , who obtained ir ( 1.6 and 2.1 @xmath11 m ) long - slit spectra in which the emission of the northern counter - jet was finally detected .
these authors note that the emission of the counterjet has intensity peaks at positions ( i. e. , distances from the source ) which approximately coincide with the knots along the southern jet . in this paper , we present new spitzer irac images of hh 34 .
these images show the southern jet and northern counterjets with comparable intensities , and with a surprising degree of symmetry .
the observations are described in section 2 . in section 3
we present an image of the central region of the hh 34 outflow , quantitatively evaluate the degree of symmetry between the jet and the counterjet , and discuss the implications of the results for the ejection mechanism that has produced the outflow .
the results are summarized in section 4 .
the observations of hh 34 are part of our original spitzer space telescope ( werner et al .
2004 ) general observer ( go ) program 3315 ( pi noriega - crespo ) obtained with both the infrared camera irac ( fazio et al .
2004 ) and the infrared photometer mips ( rieke et al . 2004 ) in march 28 , 2005 .
the data have been recovered from the spitzer legacy archive and the quality of the final images ( post basic calibrated data or post - bcd ; s18.7 products ) is outstanding , so that no further processing was required . in this study
we present the irac observations obtained in the four channels ( 1 , 2 , 3 , 4 ) = ( 3.6 , 4.5 , 5.8 & 8.0 ) covering a fov of @xmath12 ( the result of a 6@xmath136 array map with a 260 stepsize ) and with a total integration time per pixel of 30 sec . the final images are sampled with 0.6 per pixel , nearly one third of standard @xmath9 irac angular resolution .
figure 1 shows a three color image of hh 34 using channels 1 , 2 and 3 . like with other protostellar outflows
observed with irac ( see e.g. noriega - crespo et al .
2004 ; looney , tobin & kwon 2007 ; tobin et al . 2007 ;
ybarra & lada 2009 ) channel 2 recovers the strongest jet emission , since its bandpass ( 4 to 5 ) includes three relatively bright pure rotational h@xmath14 emission lines , 0 - 0 s(9 ) 4.69 , 0 - 0 s(10 ) 4.41 and 0 - 0 s(11 ) 4.18 ; the jet is well detected in channel 3 as well , where another couple of h@xmath14 lines are found ( 0 - 0 s(6 ) 6.11 and 0 - 0 s(7 ) 5.51 ) ( see de buizer & vacca 2010 ) .
in figure 2 , we show the 4.5 + 5.8 image of the region around the source of hh 34 , rotated so that the outflow axis is parallel to the ordinate .
this image shows a surprising symmetry between the southern jet ( detected in optical images ) and the northern counterjet . we have defined a 7 pixel ( @xmath15 ) wide , recangular box aligned with the hh 34 axis , within which we subtract the background emission ( defined as a linear interpolation between the pixels immediately outside of the long edges of the box ) .
the result of this background emission subtraction is shown on the right hand side of figure 2 , in which the knot structure along the hh 34 jet and counterjet is seen more clearly .
we have carried out paraboloidal fits to the emission peaks of the knots along the outflow axis , determining the positions of the peaks with an error of @xmath16 pix .
@xmath17 . through this procedure
, we obtain the positions @xmath18 and @xmath19 of the 7 knots along the jet and the counterjet ( respectively ) .
the position of the source is not so well determined , because the image of the source is partially saturated , and has a complex point spread function .
because of this , we have estimated the position of the source as the average of the coordinates of the seven knots along the jet and the counterjet . a paraboloidal fit to the emission of the region around the source actually results in a similar position ( with offsets of @xmath20 pix along and @xmath21 pix across the jet axis with respect to the average position of the ensemble of knots ) . in figure 3
, we plot the positions @xmath19 of the consecutive knots along the counterjet as a function of the positions @xmath18 of the corresponding knots along the jet .
this figure shows the remarkable symmetry ( with respect to the position of the source ) of the knots along the jet and the counterjet . in figure 3
, we also plot the offsets @xmath22 ( between the positions of the corresponding jet / counterjet knot pairs ) as a function of @xmath18 .
the first three knots have monotonically growing offsets , with @xmath23 ( ranging from @xmath24 times the measurement error , see above ) .
the four knots further away from the source generally have larger offsets , with a top value @xmath25 for the 6th knot out from the source .
the resulting @xmath26 vs. @xmath18 dependence therefore has low offsets close to the source , and increasingly large ( and more highly oscillating ) values for larger distances from the source .
these results can be interpreted as follows .
let us first assume that the jet / counterjet knot pairs are ejected at the same time @xmath27 , but with different velocities @xmath28 and @xmath29 ( for the jet and counterjet knots , respectively ) . if the knots are ballistic , at a time @xmath30 they will have positions @xmath31 and @xmath32 .
one then obtains that the jet / counterjet knot offset @xmath33 ( due to asymmetries in the jet / counterjet ejection velocities ) is given by : @xmath34 in other words , an ejection velocity asymmetry in the simultaneous ejection of a jet / counterjet knot pair results in a knot position asymmetry @xmath33 that increases linearly with distance from the source ( as the knots travel away from the source ) .
let us now assume that we have a pair of knots ejected with the same velocity @xmath28 , but at times @xmath27 and @xmath35 ( for the knot along the jet and the counterjet , respectively ) . for ballistic knots , at a time
@xmath30 they would then be at distances @xmath31 and @xmath36 from the source .
the jet / counterjet knot offset @xmath37 ( due to a time - defference @xmath38 in the ejection time ) is given by : @xmath39 in other words , the position asymmetry @xmath37 does not change as the knots travel away from the source along the jet and counterjet . for the case of jet / counterjet knot pairs ejected with a time difference @xmath38 and a velocity difference
@xmath40 , we therefore predict an asymmetry between the jet and counterjet knot positions of the form @xmath41 this relation gives the behaviour of @xmath26 as a function of distance from the source @xmath18 ( along the jet ) for a given pair of knots as they travel down the jet / counterjet axis . if we have a series of knot pairs ( 7 knot pairs in our hh 34 image ) , the values of @xmath42 and @xmath43 will not necessarily be identical for all of the knot pairs .
we can anyway attempt to fit a linear @xmath26 vs. @xmath18 dependence to the observations
. at large enough distances from the source , equation ( [ dx ] ) is a straight line which goes through the origin ( of the @xmath26 vs. @xmath18 coordinates ) , with a slope of @xmath44 .
the @xmath43 term ( see equation [ dx ] ) results in a scatter of points above and below ( depending on the sign of @xmath38 this straight line .
this scatter will also have a contribution from the distribution of the @xmath44 values over the knot pairs .
therefore , if we have a distribution of @xmath38 ( over the different knot pairs ) with a zero mean velocity ( i. e. , with similar positive and negative @xmath38 values ) , we would expect to recover from the data a straight line that still goes through the origin .
the slope of this line would represent an average of @xmath44 over the ensemble of knot pairs ( see equation [ dx ] ) .
the value of the determined error for the @xmath45 interception point of the straight line would give us an upper limit to the root mean square average of @xmath43 .
fitting a least squares linear dependence to the observed offsets , we obtain @xmath46 we therefore obtain an estimate for the average time offset of basically 0 ( as expected ) , an upper limit for its root mean square value of @xmath47 , and a velocity asymmetry with an average modulus of approximately @xmath48 .
if we consider a distance of 417 pc ( to hh 34 ) and that the knots have a plane of the sky velocity @xmath49 km s@xmath5 , we then obtain @xmath50 the maximum value of @xmath38 that we have determined in this way from the observed jet / counterjet asymmetry of hh 34 directly implies a maximum possible size for the region in which the ejection is produced .
the gas at the foot of a stellar jet is observed to have temperatures of @xmath51 k , implying a sound speed of @xmath52 km s@xmath5 .
the alvn velocity in this region is expected to have similar values . in order to synchronize the jet / counterjet ejections to a time @xmath38 , the region producing both outflows has to be able to communicate causally within this time interval .
therefore , waves propagating at a velocity @xmath53 ( this could be the alfvn or the sound velocity , both @xmath54 km s@xmath5 , see above ) should be able to travel the characteristic size @xmath55 of the jet formation region within a time @xmath38 . this condition can be written as : @xmath56 where for the second inequality we have used the value of @xmath38 derived from the observed hh 34 jet / counterjet asymmetry ( see equation [ fit2 ] ) and the estimated sound speed of 3 km s@xmath5 .
we have obtained irac images of the region around the source of the hh 34 outflow ( see figure 1 ) .
these images show the presence of a southern jet and a northern counterjet ( previously only detected in the ir long - slit spectrum presented by garca lpez et al .
the seven knots along the jet and the counterjet have a remarkable symmetry with respect to the position of the outflow source ( see figure 2 ) .
we find that if we plot the offsets @xmath22 ( where @xmath18 is the distance from the source of a knot along the jet and @xmath19 the distance to the corresponding knot along the counterjet ) as a function of the position @xmath18 of the knots , we obtain a general trend of lager @xmath26 values for increasing distances from the source ( see figure 3 ) .
this trend can be interpreted ( on the basis of a simple , ballistic knot model ) as the result of a time - offset @xmath38 ( between the ejection of the two knots in a given pair ) and of a velocity difference @xmath40 ( between the velocities of the knot pair ) . from a fit to the observed @xmath26 vs. @xmath57 relationship , we determine a velocity asymmetry of @xmath58 km s@xmath5 , and an upper limit for the time - offset of the binary ejection of @xmath59 yr .
interestingly , the velocity asymmetry estimated for the binary ejections is within a factor of @xmath9 of the estimates for the sound and alfvn speeds at the base of an outflow from a young star . actually , the estimates of the sound and alfvn speeds lie within the error bars of the @xmath40 value determined from the hh 34 jet / counterjet knot asymmetries .
it is possible to use the upper limit of @xmath60 yr to derive an estimate @xmath61 au for the size of the jet production region .
for example , if one considers models of magnetocentrifugally driven disk winds , this estimate would imply a characteristic radius for the alfvn surface ( in which the outflow becomes super - alfvnic ) of @xmath62 au .
this is an interesting constraint on models of the production of jets from young stars ( for a recent discussion of jet production models see the review of ferreira 2008 , and the papers in the book of garca & ferreira 2010 ) .
another possible interpretation is that the @xmath63 au radius ( derived from the observations ) corresponds to the outer radius of the region of the accretion disk which gives rise to the jet .
this interpretation might be more comfortable in terms of magnetocentrifugally driven disk winds , because an alfvn surface of @xmath0 au radius might be too small .
vla observations of sources of hh outflows had provided a constraint of a maximum size of @xmath64 au ( @xmath65 at the distance of orion ) on the size of the jet production region ( see , e. g. , the hh 1/2 observations of rodrguez et al . 2000 ) .
the upper limit of @xmath66 au that we have derived above for the size of the formation region of the hh 34 jet is therefore the most stringent condition yet .
this size corresponds to @xmath67 mas , a resolution that will only be approached in direct observations of an hh source by future efforts with the alma interferometer .
observations with large baseline near - ir interferometers ( such as keck - i or vlt - i ) could possibly also resolve such a region , provided that the emission is strong enough to be detected . to conclude ,
we should point out that the derived upper limit of @xmath66 au for the jet formation region is derived from a simple , ballistic approximation for the motion of the knots along the hh 34 jet and counterjet .
of course , the observed asymmetry could partially arise from deviations from a purely ballistic motion .
if this is the case , the size implied for the jet formation region would be even smaller .
this work is based in part on observations made with the _
spitzer space telescope _ which is operated by the jet propulsion laboratory , california institute of technology under nasa contract 1407 .
the work of ar and vl was supported by the conacyt grants 61547 , 101356 and 101975 . | we present a new irac , spitzer irac images of the hh 34 outflow .
these are the first images that detect both the knots along the southern jet and the northern counterjet ( the counterjet knots were only detected previously in a long slit spectrum ) .
this result removes the problem of the apparent coexistence of a large scale symmetry ( at distances of up to @xmath0 pc ) and a complete lack of symmetry close to the source ( at distances of @xmath1 cm ) for this outflow .
we present a quantitative evaluation of the newly found symmetry between the hh 34 jet and counterjet , and show that the observed degree of symmetry implies that the jet production region has a characteristic size @xmath2 au .
this is the strongest constraint yet derived for the size of the region in which hh jets are produced . |
the hubbard model on a bipartite lattice is the simplest realistic toy model for description of the electronic correlation effects in general many - electron problems with short - range interaction on such a lattice .
the model involves two effective parameters : the in - plane nearest - neighbor transfer integral @xmath0 and the effective on - site repulsion @xmath1 . despite
that it is among the mostly studied models in condensed matter and ultra - cold atom physics , except for the one - dimensional ( 1d ) bipartite lattice @xcite there is no exact solution and few controlled approximations exist for finite @xmath10 values .
one of the few exact results for the model on a general bipartite lattice is that the ground state of the repulsive half - filled hubbard model is for finite number of lattice sites and spin density @xmath12 a spin singlet @xcite .
a method to achieve further useful information about the hubbard model on a bipartite lattice such as the square lattice @xcite , honeycomb lattice @xcite , or cubic lattice @xcite involves the use of suitable systems of ultra - cold fermionic atoms .
indeed , one may expect very detailed experimental results over a wide range of parameters to be available @xcite .
preliminary results referring to the mott - hubbard insulating phase of the the hubbard model on a cubic lattice are reported in ref .
@xcite .
there is another exact result concerning the hubbard model on any bipartite lattice .
it is that in addition to the spin @xmath11 symmetry the model has a second global @xmath11 symmetry @xcite , called by some authors and in this paper @xmath13-spin symmetry @xcite .
a trivial result is that at @xmath14 the global symmetry of the hubbard model on a bipartite lattice at vanishing chemical potential and magnetic field is @xmath15 . here
the factor @xmath16 refers to the particle - hole transformation on a single spin under which the model hamiltonian is not invariant for @xmath4 and @xmath7/z_2 $ ] contains the two @xmath11 symmetries .
following work by heilmann and lieb @xcite , yang and zhang considered the most natural possibility that the @xmath17 symmetry inherited from the @xmath14 hamiltonian @xmath15 symmetry was the model global symmetry for @xmath18 @xcite .
however , a recent study of the problem by the author and collaborators reported in ref .
@xcite reveals an exact extra hidden global @xmath5 symmetry emerging for @xmath4 in addition to @xmath17 .
it is related to the @xmath4 local @xmath19 gauge symmetry of the hubbard model on a bipartite lattice with vanishing transfer integral @xcite .
such a local @xmath19 gauge symmetry becomes for finite @xmath1 and @xmath0 a group of permissible unitary transformations .
it is such that the corresponding local @xmath5 canonical transformation is not the ordinary @xmath5 gauge subgroup of electromagnetism .
instead it is a nonlinear " transformation @xcite . for @xmath4
the related new found global symmetry of the model on any bipartite lattice is larger than @xmath17 and given by @xmath20/z_2 ^ 2=[so(4)\times u(1)]/z_2=so(3)\times so(3)\times u(1)$ ] .
the factor @xmath21 ( and @xmath22 ) in @xmath7/z_2 $ ] ( and in @xmath20/z_2 ^ 2 $ ] ) imposes that @xmath23 $ ] is an integer number ( and both @xmath24 $ ] and @xmath25 $ ] are integer numbers ) . here
@xmath26 , @xmath27 , and @xmath28 are the @xmath13-spin , the spin , and the eigenvalue of the generator of the new global @xmath5 symmetry , respectively .
the latter is found in ref .
@xcite to be one half the number of rotated - electron singly occupied sites .
this refers to any of the infinite electron - rotated - electron unitary transformations of ref . @xcite , such that rotated - electron single and double occupancy are good quantum numbers for @xmath29 .
@xmath30 is then the number of rotated - electron singly occupied sites . within the present notation , @xmath31/2
$ ] and @xmath32/2 $ ] are the @xmath13-spin projection and spin projection , respectively , and @xmath33 denotes the number of lattice sites . for the bipartite 1d and square lattices considered in this paper the labeling index @xmath34 in @xmath35^d$ ] reads @xmath36 and @xmath37 , accounting for the @xmath38 and @xmath39 lattice sites , respectively .
the square and 1d lattices have spacing @xmath40 and length edge and chain length @xmath41 , respectively .
an important point is that although addition of chemical - potential and magnetic - field operator terms to the hubbard model on a bipartite lattice hamiltonian lowers its symmetry , such terms commute with it .
therefore , the global symmetry being @xmath3 implies that the set of independent rotated - electron occupancy configurations that generate the model energy and momentum eigenstates generate state representations of that global symmetry for all values of the electronic density @xmath42 and spin density @xmath43 .
it then follows that the total number of such independent representations must equal the hilbert - space dimension , @xmath44 .
the results of ref .
@xcite confirm that for the model on a bipartite lattice in its full hilbert space the number of independent representations of the group @xmath3 is indeed @xmath44 .
in contrast , the number of independent representations of the group @xmath17 is found to be smaller than the hilbert - space dimension @xmath44 .
studies of the hubbard model on a bitartie lattice that rely on its transformation properties under symmetry operations @xcite should be extended to account for the new hidden @xmath5 symmetry .
the hubbard model on the bipartite 1d lattice has an exact solution @xcite , which can be shown to be fully consistent with its extended global symmetry .
the exact solution of 1d integrable models can be reached by two different methods : the coordinate bethe ansatz ( ba ) used by bethe himself @xcite and the more algebraic operator formulation usually called inverse scattering method @xcite . for the 1d hubbard model such an algebraic formulation of the bethe states refers to the transfer matrix of the classical coupled spin model , which is its covering " @xcite .
indeed , within the inverse scattering method @xcite the central object to be diagonalized is the quantum transfer matrix rather than the underlying 1d hubbard model .
the transfer - matrix eigenvalues provide the spectrum of a set of conserved charges .
the diagonalization of the charge degrees of freedom involves a transfer matrix associated with a charge monodromy matrix @xcite .
its off - diagonal entries are some of the creation and annihilation fields whose application onto a suitable vacuum generates the charge degrees of freedom of the energy eigenstates configurations .
the solution of the spin degrees of freedom involves the diagonalization of the auxiliary transfer matrix associated with the spin monodromy matrix @xcite .
again , the off - diagonal entries of that matrix play the role of the creation and annihilation fields that generate the spin degrees of freedom of the energy eigenstates configurations .
the global symmetry of a solvable model is explicit in the algebraic operator formulation of its solution .
this is why the solution of the 1d hubbard model by the algebraic inverse scattering method @xcite was achieved only thirty years after that of the coordinate ba @xcite .
indeed , it was expected that the charge and spin monodromy matrices of the former method had the same traditional faddeev - zamolodchikov abcd form , found previously for the related 1d isotropic heinsenberg model @xcite .
such an expectation was that consistent with the occurrence of a spin @xmath11 symmetry and a @xmath13-spin charge @xmath11 symmetry associated with a global @xmath7/z_2 $ ] symmetry @xcite . if that was the whole global symmetry of the 1d hubbard model , the charge and spin monodromy matrices would have the same abcd form .
consistently with the recently found model extended global symmetry , all tentative schemes using charge and spin monodromy matrices of the same abcd form failed to achieve the ba equations obtained by means of the coordinate ba @xcite . as discussed in appendix a , the problem was solved by martins and ramos , who used an appropriate representation of the charge and spin monodromy matrices , which allows for possible _ hidden symmetries
indeed , the structure of the charge and spin monodromy matrices introduced by these authors is able to distinguish creation and annihilation fields as well as possible hidden symmetries .
that for @xmath18 the 1d hubbard model spin and charge degrees of freedom are associated with the @xmath11 and @xmath45 symmetries , respectively , of @xmath46/z_2 ^ 2=so(3)\times so(3)\times u(1)$ ] rather than merely with two @xmath11 symmetries , is behind the different abcd and abcdf forms found in ref .
@xcite for the inverse - scattering method ba solution spin and charge monodromy matrices , respectively : due to the extra @xmath5 symmetry in @xmath45 , the latter matrix is larger than the former and involves more fields . although the model on bipartite lattices other than the 1d lattice has no exact solution to be related to its global symmetry , that for @xmath18 such a symmetry is @xmath3
is expected to have important physical consequences .
one of the goals of this paper is the introduction of a general quantum - object description for the hubbard model on the bipartite square lattice to be used in ref . @xcite and elsewhere in the search of such physical consequences . to meet that goal , the description introduced in this paper accounts for a set of independent rotated - electron occupancy configurations that generate the model state representations of the global @xmath3 symmetry for all values of the electronic density @xmath42 and spin density @xmath43 . for the 1d hubbard model
such a complete set of states associated with the operator description introduced here are energy eigenstates and refer to the exact solution quantum numbers and set of conserving quantities . due to the square - lattice hubbard model lack of integrability and corresponding lack of an infinite number of conservation laws , for it only in the one- and two - electron subspace introduced in section v
are the above states energy eigenstates .
fortunately , such model in that subspace describes the important one- and two - electron physics . furthermore
, one of the motivations for our study focusing onto the hubbard model on the square lattice follows from that model with a small three - dimensional ( 3d ) uniaxial anisotropy perturbation being the simplest candidate toy model for describing the effects of electronics correlations in the high-@xmath47 superconductors @xcite .
the first step of our program involves an electron - rotated - electron unitary transformation of the type considered in ref .
a property specific to our transformation is that the @xmath18 energy eigenstates can be generated from suitable chosen @xmath48 energy eigenstates upon application onto the latter states of the corresponding electron - rotated - electron unitary operator .
the related set of state representations of the group @xmath3 that emerge from our description are energy eigenstates yet are generated by exactly the same electron - rotated - electron unitary transformation from corresponding @xmath48 states .
their rotated electron occupancy configurations are simpler to describe in terms of three types of elementary objects directly related to the rotated electrons whose numbers and designations are : @xmath49 spin-@xmath9 spinons , @xmath50 $ ] @xmath13-spin-@xmath9 @xmath13-spinons , and @xmath51 spin - less and @xmath13-spin - less charge @xmath52 fermions .
the latter live on a lattice with @xmath53 $ ] sites identical to the original lattice . here
$ ] gives the number of @xmath52 fermion holes .
the relation of such objects to the rotated electrons is as follows .
the @xmath49 spin-@xmath9 spinons describe the spin degrees of freedom of the @xmath30 rotated electrons that singly occupy sites .
the charge degrees of freedom of such rotated electrons are described by the @xmath55 @xmath52 fermions .
the @xmath50 $ ] @xmath13-spin-@xmath9 @xmath13-spinons describe the @xmath13-spin degrees of freedom of the @xmath56 $ ] sites doubly occupied and unoccupied by rotated electrons .
specifically , the @xmath13-spinons of @xmath13-spin projection @xmath57 and @xmath58 refer to the sites doubly occupied and unoccupied , respectively , by rotated electrons .
the remaining degrees of freedom of such rotated - electron occupancy configurations are described by the @xmath54 $ ] @xmath52 fermion holes .
the expression of the rotated - electron operators in terms of these three elementary objects is for @xmath18 identical to that of the electron operators for large @xmath1 values in terms of the three objects obtained from the exact transformation for separation of spin-@xmath9 fermions without constraints considered in ref .
@xcite .
interestingly , @xmath13-spinon ( and spinons ) that are not invariant under the electron - rotated - electron unitary transformation considered in this paper have @xmath13-spin @xmath9 ( and spin @xmath9 ) but are confined within @xmath13-spin - neutral ( and spin - neutral ) @xmath59-@xmath13-spinon ( and @xmath59-spinon ) composite @xmath60 fermions ( and @xmath61 fermions ) . here
@xmath62 is the number of confined @xmath13-spinon ( and spinon ) pairs .
we emphasize that by confinement " is meant here that the spinons and @xmath13-spinons are bound and anti - bound within such composite @xmath61 and @xmath60 fermions , respectively , alike for instance protons and neutrons are bound within the nucleus .
whether the potential associated with such a behavior is in some limit confining remains an interesting open question . in turn ,
a well - defined number of deconfined spinons and deconfined @xmath13-spinons are invariant under the electron - rotated - electron unitary transformation . confined and deconfined
spinons play an important role in the concept of deconfined quantum criticality @xcite .
the relation of the physical consequences of the model extended global symmetry to such a concept is one of the interesting issues raised by our general description to be investigated elsewhere . in this paper
often we discuss the use of our description for the model on a 1d lattice as well .
indeed our quantum - object description also applies to 1d .
however , due to the integrability of the 1d model associated with its exact solution @xcite , concerning some properties it leads to a different physics .
nonetheless there are also some common properties .
one of the procedures used to control the validity of the approximations used in the construction of our description profits from identifying the common proprieties of the model on the square and 1d lattices .
some of these common properties are related to the model commuting with the generators of the group @xmath3 both for the square and 1d bipartite lattices @xcite . as confirmed in appendix a , for the 1d hubbard model the states generated by the simple momentum occupancy configurations of the objects of our description are exact energy eigenstates for the whole hilbert space and all finite values of the onsite repulsion @xmath1 . for the hubbard model on the square lattice
such occupancy configurations generate momentum eigenstates that are not in general energy eigenstates .
fortunately , they are so for that model in the one- and two - electron subspace defined in section v , which refers to the square - lattice quantum liquid introduced in that section and further studied in ref .
@xcite .
the paper is organized as follows .
a uniquely defined rotated - electron description for the hubbrad model on the square lattice is the subject of section ii . in section
iii the @xmath52 fermion , @xmath13-spin-@xmath9 @xmath13-spinon , and spin-@xmath9 spinon and corresponding @xmath52 , @xmath13-spin , and spin effective lattices are introduced .
the vacua of the theory , the transformation laws under the electron - rotated - electron unitary transformation of such objects , and the subspaces they refer to are issues also addressed in that section .
the composite @xmath63 bond particles and @xmath63 fermions , corresponding @xmath63 effective lattices , ground - state occupancy configurations , and a complete set of momentum eigenstates are the problems studied in section iv .
the one- and two - electron electron subspace and the corresponding square - lattice quantum liquid are the subjects of section v. in section vi the local @xmath64 fermion operators and related @xmath64 bond - particle operators are expressed in terms of elementary spinon operators .
section vii contains the concluding remarks . in appendix
a it is confirmed that the general operator description introduced in this paper for the hubbard model on the square and 1d lattices is consistent with the exact solution of the 1d problem .
for instance , it is confirmed that the discrete momentum values of the @xmath52 fermions and composite @xmath63 fermions are good quantum numbers .
furthermore , in that appendix the relation of the creation and annihilation fields of the charge abcdf algebra @xcite and more traditional spin abcd faddeev - zamolodchikov algebra @xcite of the algebraic formulation of the 1d exact solution of ref .
@xcite to the @xmath52 and @xmath63 fermion operators is discussed .
the consistency between the two corresponding operational representations is confirmed . moreover , that the whole physics can be extracted from the hubbard model in the lowest - weight state ( lws ) subspace spanned by the @xmath13-spin and spin lwss is the subject addressed in appendix b. in appendix c additional side information on the one- and two - electron subspace is provided .
finally , appendix d presents some technical results on the @xmath64 bond - particle operators algebra .
the hubbard model on a square ( or 1d ) lattice with a very large number @xmath33 of sites reads , @xmath65/2 \ , ; \hspace{0.25 cm } \hat{t } = -\sum_{\langle\vec{r}_j\vec{r}_{j'}\rangle}\sum_{\sigma}[c_{\vec{r}_j,\sigma}^{\dag}\,c_{\vec{r}_{j'},\sigma}+h.c . ]
\ , ; \hspace{0.25 cm } { \hat{q } } = \sum_{j=1}^{n_a^d}\sum_{\sigma = \uparrow , \downarrow}\,\hat{n}_{\vec{r}_j,\sigma}\,(1- \hat{n}_{\vec{r}_j , -\sigma } ) \ , .
\label{h}\ ] ] periodic boundary conditions and torus periodic boundary conditions are considered for the 1d lattice for which @xmath36 and the square lattice for which @xmath37 , respectively . moreover , in eq .
( [ h ] ) @xmath0 is the nearest - neighbor transfer integral , @xmath66 is the kinetic - energy operator in units of @xmath0 , and @xmath67 is the operator that counts the number of electron singly occupied sites .
hence the operator @xmath68/2 $ ] counts the number of electron doubly occupied sites .
moreover , @xmath69 where @xmath70 ( and @xmath71 ) for @xmath72 ( and @xmath73 ) , @xmath74 , and @xmath75 . the kinetic - energy operator @xmath66 can be expressed in terms of the operators , @xmath76 \ , , \nonumber \\ \hat{t}_{+1 } & = & -\sum_{\langle\vec{r}_j\vec{r}_{j'}\rangle}\sum_{\sigma } \hat{n}_{\vec{r}_{j},-\sigma}\,c_{\vec{r}_j,\sigma}^{\dag}\,c_{\vec{r}_{j'},\sigma}\,(1-\hat{n}_{\vec{r}_{j'},-\sigma } ) \ , , \nonumber \\ \hat{t}_{-1 } & = & -\sum_{\langle\vec{r}_j\vec{r}_{j'}\rangle}\sum_{\sigma } ( 1-\hat{n}_{\vec{r}_{j},-\sigma})\,c_{\vec{r}_j,\sigma}^{\dag}\ , c_{\vec{r}_{j'},\sigma}\,\hat{n}_{\vec{r}_{j'},-\sigma } \ , , \label{t - op}\end{aligned}\ ] ] as @xmath77 .
these three kinetic operators play an important role in the physics .
the operator @xmath78 does not change electron double occupancy whereas the operators @xmath79 and @xmath80 change it by @xmath81 and @xmath82 , respectively .
we focus our attention onto ground states with hole concentration @xmath83/n_a^d\geq 0 $ ] and spin density @xmath84/n_a^d=0 $ ] and their excited states .
we are particularly interested in the _ lws subspace _ spanned by the lwss of both the @xmath13-spin and spin algebras .
such energy eigenstates refer to values of @xmath85 and @xmath86 such that @xmath87 for @xmath88 .
the off - diagonal generators that generate the non - lwss from the lwss commute with the operator @xmath89 @xcite .
thus the whole physics can be extracted from the model ( [ h ] ) in the lws subspace , as confirmed in appendix b. the studies of ref .
@xcite consider unitary operators @xmath90 and corresponding rotated - electron operators , @xmath91 those are such that rotated - electron single and double occupancy are good quantum numbers for @xmath8 .
the global @xmath5 symmetry generator @xmath92 of eigenvalue @xmath28 reads @xcite , @xmath93 it follows that @xmath94 where @xmath95 and the operator @xmath67 is given in eq .
( [ h ] ) . as mentioned in the previous section ,
@xmath30 is the number of rotated - electron singly occupied sites .
most choices of the unitary operators @xmath89 correspond to choices of @xmath48 sets @xmath96 of @xmath44 energy eigenstates such that the states @xmath97 are not energy and momentum eigenstates for finite @xmath10 values yet belong to a subspace with fixed and well - defined values of @xmath28 , @xmath26 , and @xmath27 .
let @xmath98 be a complete set of @xmath44 energy , momentum , @xmath13-spin , @xmath13-spin projection , spin , and spin - projection eigenstates for @xmath8 . in the limit @xmath48 such states correspond to one of the many choices of sets @xmath96 of @xmath44 @xmath10-infinite energy eigenstates .
both the sets of states @xmath98 and @xmath99 , respectively , are complete and the model hilbert space is the same for all @xmath8 values considered here .
hence it follows from basic quantum mechanics hilbert - space and operator properties that for this choice there exists exactly one unitary operator @xmath100 such that @xmath101 .
here we consider such a unitary operator and corresponding generator @xmath94 given in eq .
( [ or - ope ] ) and rotated - electron operators provided in eq .
( [ rotated - operators ] ) .
the states @xmath101 ( one for each value of @xmath8 ) that are generated from the same initial state @xmath102 belong to the same _
@xmath103 tower_. a complete set @xmath104 of related momentum eigenstates @xmath105 is considered below in section iv - e .
such states are generated by occupancy configurations of the quantum objects of the general description introduced in this paper .
the unitary operator @xmath106 appearing in the general expression of such states is that which also appears in the general expression @xmath101 of the above energy and momentum eigenstates .
however the states @xmath107 are not in general energy eigenstates of the hubbard model on the square lattice .
the interest of the states @xmath105 is that those contained in the one- and two - electron subspace defined in section v are both momentum and energy eigenstates of that model .
fortunately , the general description introduced in this paper is physically most useful and important for the hubbard model on the square lattice in that subspace , which refers to the square - lattice quantum liquid further investigated in ref .
@xcite . alike in ref .
@xcite , we associate with any operator @xmath108 an operator @xmath109 that has the same expression in terms of rotated - electron creation and annihilation operators as @xmath108 in terms of electron creation and annihilation operators , respectively .
our convention is that marks placed over letters being a caret or a tilde denote operators .
( an exception are the electron operators of eq .
( [ rotated - operators ] ) , which we denote by @xmath110 and @xmath111 rather than by @xmath112 and @xmath113 , respectively . )
any operator @xmath108 can then be written in terms of rotated - electron creation and annihilation operators as , @xmath114 + { 1\over 2}\,[[{\tilde{o}},{\hat{s}}\,],{\hat{s}}\ , ] + ... = { \tilde{o}}+ [ { \tilde{o}},{\tilde{s}}\ , ] + { 1\over 2}\,[[{\tilde{o}},{\tilde{s}}\,],{\tilde{s}}\ , ] + ... \ , , \nonumber \\ { \hat{s } } & = & -{t\over u}\,\left[\hat{t}_{+1 } -\hat{t}_{-1}\right ] + { \cal{o } } ( t^2/u^2 ) \ , ; \hspace{0.25 cm } { \tilde{s } } = -{t\over u}\,\left[\tilde{t}_{+1 } -\tilde{t}_{-1}\right ] + { \cal{o } } ( t^2/u^2 ) \ , .
\label{oor}\end{aligned}\ ] ] the operator @xmath115 appearing in this equation is related to the unitary operator as @xmath116 and @xmath117 .
although the general expression of @xmath118 remains unknown , an exact result is that it involves only the kinetic operators @xmath78 , @xmath79 , and @xmath80 of eq . ( [ t - op ] ) and numerical @xmath10 dependent coefficients @xcite . for @xmath8 that expression can be expanded in a series of @xmath119 .
the corresponding first - order term has the universal form given in eq .
( [ oor ] ) . to arrive to the expression of @xmath108 in terms of the operator @xmath120
also given in eq .
( [ oor ] ) , the property that the operator @xmath121 commutes with itself is used .
it implies that @xmath122 and @xmath123 .
hence both the operators @xmath121 and @xmath118 have the same expression in terms of electron and rotated - electron creation and annihilation operators .
this justifies why the expansion @xmath124 + { \cal{o } } ( t^2/u^2)$ ] given in that equation for the operator @xmath120 has the same form as that of @xmath118 .
the higher - order terms of the operator @xmath120 expression can be written as a product of operator factors .
their expressions involve the rotated kinetic operators @xmath125 , @xmath126 , and @xmath127 . the full expression of the operator @xmath128 can for @xmath8 be written as @xmath129 . here
@xmath130 corresponds to the operator @xmath131 at @xmath132 defined in eq .
( 61 ) of ref .
moreover , @xmath133 has the general form provided in eq .
( 64 ) of that paper .
the unitary operator @xmath134 considered here corresponds to exactly one choice of the coefficients @xmath135 of that equation . here
the index @xmath136 refers to the number of rotated - electron doubly occupied sites .
the problem of finding the explicit form of the operators @xmath137 and @xmath123 is equivalent to finding all coefficients @xmath135 associated with the electron - rotated - electron unitary transformation as defined in this paper . for finite @xmath10 values the hamiltonian @xmath138 of eq .
( [ h ] ) does not commute with the unitary operator @xmath117 .
hence when expressed in terms of the rotated - electron creation and annihilation operators of eq .
( [ rotated - operators ] ) it has an infinite number of terms . according to eq .
( [ oor ] ) it reads , @xmath139 + { 1\over 2}\,[[{\tilde{h}},{\tilde{s}}\,],{\tilde{s}}\ , ] + ...
\label{hhr}\ ] ] the commutator @xmath140 $ ] does not vanish except for @xmath48 so that @xmath141 for finite values of @xmath10 .
fortunately , for approximately @xmath142 , out of the infinite terms on the right - hand - side of eq .
( [ hhr ] ) only the first few hamiltonian terms play an active role in the physics of the hubbard model on the square lattice in the one- and two - electron subspace @xcite . alike
the operator @xmath120 , the hamiltonian expression in terms of rotated - electron operators ( [ hhr ] ) can be expanded in a series of @xmath119 .
its terms generated up to fourth order in @xmath119 are within a unitary transformation the equivalent to the @xmath143 model with ring exchange and various correlated hoppings @xcite . furthermore and in accordance to a general theorem proved in ref .
@xcite , at half filling the terms of the hamiltonian ( [ hhr ] ) expansion in @xmath119 with odd powers in @xmath0 vanish due to the particle - hole symmetry and the resulting invariance of the spectrum under @xmath144 .
in turn , for finite hole concentration @xmath145 the expansion in powers of @xmath119 of the hamiltonian ( [ hhr ] ) involves terms with odd powers in @xmath0 , absent at @xmath146 .
this is consistent with the effects of increasing @xmath10 being often different at @xmath146 and for @xmath145 @xcite . in ref .
@xcite it is found that , in contrast to the hamiltonian , the three components of the momentum operator @xmath147 , three generators of the spin @xmath11 symmetry , and three generators of the @xmath13-spin @xmath11 symmetry commute with the electron - rotated - electron unitary operator @xmath148 .
this also holds for the specific choice of that operator associated with the rotated - electron description introduced in this paper .
hence the above operators have the same expression in terms of electron and rotated - electron creation and annihilation operators , so that the momentum operator reads , @xmath149 furthermore , the above - mentioned six generators are given by , @xmath150 \ , ; \hspace{0.25 cm } { \hat{s}}_s^{x_3 } = -{1\over 2}[{\hat{n}}_{\uparrow}- { \hat{n}}_{\downarrow } ] \ , , \nonumber \\ { \hat{s } } _ { \eta}^{\dag } & = & \sum_{j=1}^{n_a^d}e^{i\vec{\pi}\cdot\vec{r}_j}\,c_{\vec{r}_j,\downarrow}^{\dag}\ , c_{\vec{r}_j,\uparrow}^{\dag } = \sum_{j=1}^{n_a^d}e^{i\vec{\pi}\cdot\vec{r}_j}\,{\tilde{c}}_{\vec{r}_j,\downarrow}^{\dag}\ , { \tilde{c}}_{\vec{r}_j,\uparrow}^{\dag } \ , ; \hspace{0.25 cm } { \hat{s}}_{\eta } = \sum_{j=1}^{n_a^d}e^{-i\vec{\pi}\cdot\vec{r}_j}\,c_{\vec{r}_j,\uparrow}\,c_{\vec{r}_j,\downarrow } = \sum_{j=1}^{n_a^d}e^{-i\vec{\pi}\cdot\vec{r}_j}\,{\tilde{c}}_{\vec{r}_j,\uparrow}\,{\tilde{c}}_{\vec{r}_j,\downarrow } \ , , \nonumber \\ { \hat{s}}_s^{\dag } & = & \sum_{j=1}^{n_a^d}\,c_{\vec{r}_j,\downarrow}^{\dag}\,c_{\vec{r}_j,\uparrow } = \sum_{j=1}^{n_a^d}\,{\tilde{c}}_{\vec{r}_j,\downarrow}^{\dag}\,{\tilde{c}}_{\vec{r}_j,\uparrow } \ , ; \hspace{0.25 cm } { \hat{s}}_s = \sum_{j=1}^{n_a^d}c_{\vec{r}_j,\uparrow}^{\dag}\ , c_{\vec{r}_j,\downarrow}=\sum_{j=1}^{n_a^d}{\tilde{c}}_{\vec{r}_j,\uparrow}^{\dag}\ , { \tilde{c}}_{j,\downarrow } \ , .
\label{scs}\end{aligned}\ ] ] here for the model on the square ( and 1d ) lattice the vector @xmath151 has cartesian components @xmath152 $ ] ( and component @xmath153 ) .
in contrast , except in the @xmath48 limit the generator @xmath92 of the charge independent @xmath5 symmetry given in eq .
( [ or - ope ] ) does not commute with the unitary operator @xmath121 .
this is behind the hidden character of such a symmetry . on the contrary of the hamiltonian
, that generator has a complicated expression in terms of electron creation and annihilation operators and a simple expression given in that equation in terms of rotated - electron creation and annihilation operators .
the operator of eq .
( [ or - ope ] ) plus the six operators provided in eq .
( [ scs ] ) are the seven generators of the group @xmath154/z_2=so(3)\times so(3)\times u(1)$ ] associated with the global symmetry of the hamiltonian ( [ h ] ) .
the rotated electrons as defined in this paper are not the ultimate objects whose simple occupancy configurations generate the state representations of the group @xmath154/z_2=so(3)\times so(3)\times u(1)$ ] associated with the global symmetry of the model ( [ h ] ) .
the rotated - electron occupancy configurations that generate such states are naturally expressed in terms of those of three basic objects : @xmath52 fermions , spinons , and @xmath13-spinons . the local @xmath52 fermion annihilation operator @xmath155 and corresponding creation operator @xmath156 are constructed in terms of the rotated - electron operators of eq . ( [ rotated - operators ] ) .
the latter operator reads , @xmath157 in eq .
( [ fc+ ] ) we have introduced as well the corresponding @xmath52 fermion momentum - dependent operators and @xmath158 is @xmath159 depending on which sub - lattice site @xmath160 is on .
( for the 1d lattice that phase factor can be written as @xmath161 . )
the @xmath52 momentum band is studied in ref .
@xcite and has the same shape and momentum area as the electronic first - brillouin zone . consistently with the @xmath162 expression ( [ fc+ ] ) , the three spinon local operators @xmath163 and the three @xmath13-spinon local operators @xmath164 such that @xmath165 , respectively , have for @xmath8 the following expression in terms of rotated - electron operators , @xmath166 here @xmath167 where @xmath168 denotes the cartesian coordinates of the operators @xmath169 with @xmath170 and the the rotated quasi - spin operators read , @xmath171 in addition , @xmath172 is the @xmath52 fermion local density operator . since the electron - rotated - electron transformation generated by the operator @xmath89 is unitary , the operators @xmath173 and @xmath174 of eq .
( [ rotated - operators ] ) have the same anticommutation relations as @xmath110 and @xmath111 , respectively .
straightforward manipulations based on eqs .
( [ fc+])-([rotated - quasi - spin ] ) then lead to the following algebra for the @xmath52 fermion operators , @xmath175 and @xmath52 fermion operators and rotated quasi - spin operators , @xmath176 = [ f_{\vec{r}_j , c}\ , , q^l_{\vec{r}_{j ' } } ] = 0 \ , .
\label{albegra - cf - s - h}\ ] ] in turn , the rotated quasi - spin operators @xmath177 , @xmath178 , and @xmath179 and the related operators @xmath167 obey the following algebra , @xmath180 = i\,\delta_{j , j'}\sum_{p '' } \epsilon_{pp'p''}\,q^{x_{p''}}_{\vec{r}_j } \ , ; \hspace{0.15 cm } p , p',p''=1,2,3 \ , , \label{albegra - s - h}\ ] ] @xmath181 @xmath182 = \delta_{j , j'}\,q^{x_3}_{\vec{r}_j } \ , ; \hspace{0.50 cm } [ q^{\pm}_{\vec{r}_j},q^{\pm}_{\vec{r}_{j'}}]=0 \ , .
\label{albegra - q - com}\ ] ] hence the operators @xmath183 anticommute on the same site and commute on different sites .
the same applies to three spinon operators @xmath163 and three @xmath13-spinon operators @xmath164 , respectively , whose expressions are given in eq .
( [ sir - pir ] ) .
the relations provided in eqs .
( [ albegra - cf])-([albegra - q - com ] ) confirm that the @xmath52 fermions associated with the hidden global @xmath5 symmetry are @xmath13-spinless and spinless fermionic objects .
they are consistent as well with the spinons and @xmath13-spinons being spin-@xmath9 and @xmath13-spin-@xmath9 objects , respectively , whose local operators obey the usual corresponding @xmath11 algebras . on inverting the relations given in eqs .
( [ fc+ ] ) and ( [ rotated - quasi - spin ] ) , the rotated - electron creation and/or annihilation operators of eq .
( [ rotated - operators ] ) are written in terms of the operators of the @xmath52 fermions and rotated quasi - spin operators .
for the lws subspace considered here this leads to , @xmath184 for @xmath48 the rotated electrons become electrons .
in that limit the @xmath52 fermion creation operators become the quasicharge annihilation operators of ref . @xcite and
the spinon and @xmath13-spinon operators become the local spin and pseudospin operators , respectively , of that reference .
consistently , in that limit eqs .
( [ fc+])-([c - up - c - down ] ) are equivalent to eqs .
( 1)-(3 ) of ref .
@xcite with the rotated - electron operators replaced by the corresponding electron operators and the @xmath52 fermion creation operator @xmath162 replaced by the quasicharge annihilation operator @xmath185 . since the transformation considered in ref .
@xcite does not introduce hilbert - space constraints , suitable occupancy configurations of the objects associated with the local quasicharge , spin , and pseudospin operators generate a complete set of @xmath48 states @xmath96 . in section
iv - e a corresponding complete set of finite-@xmath10 states of the form @xmath105 is constructed .
those are both state representations of the model global @xmath3 symmetry and momentum eigenstates .
in general the energy and momentum eigenstates @xmath101 of the hubbard model on the square lattice considered in section ii are a superposition of a sub - set of states @xmath107 with the same momentum eigenvalue .
it follows from the results of ref .
@xcite that the @xmath13-spin @xmath11 and spin @xmath11 state representations correspond to the @xmath13-spin and spin degrees of freedom of independent rotated - electron occupancy configurations of @xmath56 $ ] sites and @xmath30 sites , respectively , of the original lattice .
this applies to any choice of electron - rotated - electron unitary transformation and thus applies as well to that associated with our description . in turn , the state representations associated with the new - found hidden @xmath5 symmetry of the global @xmath6/z_2 ^ 2 $ ] symmetry refer to the relative positions in the original lattice of the sites involved in each of these two types of configurations . in the present @xmath33 limit it is useful to introduce the following numbers , @xmath186 such that @xmath187 . here
the integer numbers @xmath188 and @xmath189 are the number of sites of the original lattice singly occupied by rotated electrons and unoccupied plus doubly occupied by rotated electrons , respectively .
below such numbers are found to play the role of number of sites of a @xmath13-spin and spin effective lattice , respectively . for the @xmath37 square lattice
the number @xmath190 is chosen so that the number @xmath191 of sites in a row or column is an integer .
however , the designations @xmath188 and @xmath189 do not imply that the corresponding numbers @xmath192 and @xmath193 are integers .
in general they are not integers . for finite values of @xmath194 and @xmath195 and @xmath196
we use for the numbers @xmath192 and @xmath193 of sites in a row and column of the @xmath13-spin and spin effective square lattices , respectively , the corresponding closest integer numbers .
importantly , the numbers given in eq .
( [ na - eta - s ] ) are fully controlled by the eigenvalue @xmath28 of the generator ( [ or - ope ] ) of the hidden global @xmath5 symmetry .
this confirms the important role plaid by such a symmetry in the general description introduced in this paper .
the degrees of freedom of the rotated - electron occupancy configurations of each of the sets of @xmath197 $ ] and @xmath198 sites of the original lattice that generate the @xmath26 , @xmath199 , @xmath27 , @xmath200 , @xmath28 , and momentum eigenstates studied in section iv , which refer to state representations of the model global @xmath3 symmetry , naturally separate as follows : \i ) the occupancy configurations of the @xmath52 fermions associated with the operators @xmath156 of eq .
( [ fc+ ] ) where @xmath201 correspond to the state representations of the hidden global @xmath5 symmetry found in ref . @xcite .
such @xmath52 fermions live on the @xmath52 effective lattice .
it is identical to the original lattice .
its occupancies are related to those of the rotated electrons : the number of @xmath52 fermion occupied and unoccupied sites is given by @xmath202 and @xmath203 $ ] , respectively .
indeed , the @xmath52 fermions occupy the sites singly occupied by the rotated electrons . in turn , the rotated - electron doubly - occupied and unoccupied sites are those unoccupied by the @xmath52 fermions . hence the @xmath52 fermion occupancy configurations describe the relative positions in the original lattice of the @xmath197 $ ] sites of the _ @xmath13-spin effective lattice _ and @xmath198 sites of the _ spin effective lattice_. \ii ) the remaining degrees of freedom of rotated - electron occupancies of the sets of @xmath197 $ ] and
@xmath198 original - lattice sites correspond to the occupancy configurations associated with the @xmath13-spin @xmath11 symmetry and spin @xmath11 symmetry state representations , respectively . the occupancy configurations of the @xmath197 $ ] sites of the @xmath13-spin effective lattice and set of @xmath198 sites of the spin effective lattice are independent .
the former configurations refer to the operators @xmath164 of eq .
( [ sir - pir ] ) , which act only onto the @xmath197 $ ] sites of the @xmath13-spin effective lattice .
the latter configurations correspond to the operators @xmath163 given in the same equation , which act onto the @xmath198 sites of the spin effective lattice .
this is assured by the operators @xmath204 and @xmath205 in their expressions provided in that equation , which play the role of projectors onto the @xmath13-spin and spin effective lattice , respectively .
for @xmath8 the degrees of freedom of each rotated - electron singly occupied site then separate into a spin - less @xmath52 fermion carrying the electronic charge and a spin - down or spin - up spinon .
furthermore , the degrees of freedom of each rotated - electron doubly - occupied or unoccupied site separate into a @xmath13-spin - less _ @xmath52 fermion hole _ and a @xmath13-spin - down or @xmath13-spin - up @xmath13-spinon , respectively .
the @xmath13-spin - down or @xmath13-spin - up @xmath13-spinon refers to the @xmath13-spin degrees of freedom of a rotated - electron doubly - occupied or unoccupied site , respectively , of the original lattice .
hence a key feature of our description is that for @xmath8 its quantum objects correspond to rotated - electron configurations whose numbers of spin - down and spin - up single occupied sites , double occupied sites , and unoccupied sites are good quantum numbers .
this is in contrast to descriptions in terms of electron configurations , whose numbers of spin - down and spin - up single occupied sites , double occupied sites , and unoccupied sites are good quantum numbers only for @xmath206 @xcite .
the transformation laws under the electron - rotated - electron unitary transformation of the @xmath13-spinons ( and spinons ) as defined in this paper play a major role in the description of the @xmath13-spin ( and spin ) @xmath11 state representations in terms of the occupancy configurations of the @xmath197 $ ] sites of the @xmath13-spin effective lattice ( and set of @xmath198 sites of the spin effective lattice ) .
indeed , a well - defined number of @xmath13-spinons ( and spinons ) remain invariant under that unitary transformation . those are called deconfined @xmath207 @xmath13-spinons ( and deconfined @xmath207 spinons ) . as further discussed below , they play the role of unoccupied sites of the @xmath13-spin ( and spin ) effective lattice .
the values of the numbers @xmath208 of deconfined @xmath207 @xmath13-spinons and @xmath209 of deconfined @xmath207 spinons are fully controlled by the @xmath13-spin @xmath26 and @xmath13-spin projection @xmath210 and spin @xmath211 and spin projection @xmath212 , respectively , of the state under consideration as follows , @xmath213=2s_{\alpha } \ , ; \hspace{0.25 cm } m^{de}_{\alpha,\pm 1/2 } = [ s_{\alpha}\mp s_{\alpha}^{x_3 } ] \ , ; \hspace{0.25 cm } \alpha = \eta \ , , s \ , .
\label{l - l}\ ] ] the invariance of such deconfined @xmath13-spinons ( and deconfined spinons ) stems from the off diagonal generators of the @xmath13-spin ( and spin ) algebra , which flip their @xmath13-spin ( and spin ) , commuting with the unitary operator @xmath89 .
this justifies why such generators have for @xmath8 the same expressions in terms of electron and rotated - electron operators , as given in eq .
( [ scs ] ) .
it follows that the number of sites of the @xmath13-spin ( @xmath214 ) and spin ( @xmath215 ) effective lattice can be written as , @xmath216 = [ m^{de}_{\alpha } + m^{co}_{\alpha } ] = m_{\alpha } \ , ; \hspace{0.25 cm } \alpha = \eta \ , , s \ , .
\label{2s-2c}\ ] ] as justified below , here @xmath217 is the confined @xmath13-spinon ( @xmath214 ) or confined spinon ( @xmath215 ) even number .
the @xmath13-spin or spin degrees of freedom of the occupancy configurations in a set of @xmath218 sites out of the whole set of @xmath219 sites have for @xmath8 the same form in terms of electrons and rotated electrons . in turn , for finite values of @xmath10 and spin density @xmath220 the corresponding @xmath52 fermion occupancy configurations that store the information on the relative positions in the original lattice of the @xmath221 sites of the spin effective lattice and @xmath222 $ ] sites of the @xmath13-spin effective lattice are not invariant under the electron - rotated - electron unitary transformation .
such an invariance of the @xmath13-spin degrees of freedom of the above occupancy configurations implies that in each state representation there are exactly @xmath223=m^{de}_{\eta}$ ] sites such that @xmath224 sites are doubly occupied and @xmath225 sites are unoccupied both by electrons and rotated electrons .
furthermore , the invariance of the spin degrees of freedom of the sites singly occupied by rotated electrons implies that there are exactly @xmath226=m^{de}_{s}$ ] sites of the original lattice such that @xmath227 sites are singly occupied both for spin - down electrons and spin - down rotated electrons and @xmath228 sites are singly occupied both for spin - up electrons and spin - up rotated electrons .
the state representation is in general a superposition of such occupancy configurations whose positions of the @xmath229=[m^{de}_{\eta}+m^{de}_{s}]$ ] sites are different . in turn , out of the set of @xmath230 $ ] sites of the original lattice left over , a set of @xmath231 sites are unoccupied by rotated electrons , a set of @xmath231 sites are doubly occupied by rotated electrons , a set of @xmath232 sites are singly occupied by spin - up rotated electrons , and a set of @xmath232 sites are singly occupied by spin - down rotated electrons .
however , in terms of electrons that set of @xmath230 $ ] sites of the original lattice has for finite @xmath10 values very involved occupancies .
indeed for electrons and except for @xmath48 singly and doubly occupancy are not good quantum numbers and thus are not conserved .
the important point brought about by our description is that due to the transformation laws under the electron - rotated - electron unitary transformation this refers only to a sub - set of @xmath230 $ ] sites out of the @xmath233 sites of the original lattice .
indeed for the remaining @xmath229=[m^{de}_{\eta}+m^{de}_{s}]$ ] sites singly and doubly occupancy are good quantum numbers both for electrons and rotated electrons .
the site numbers @xmath234 and @xmath235 are good quantum numbers given by , @xmath236 = [ n_a^d - 2s_c - 2s_{\eta } ] \ , ; \hspace{0.25 cm } m^{co}_{s } = [ n_{a_{s}}^d - 2s_{s } ] = [ 2s_c - 2s_{s } ] \ , .
\label{c - c}\ ] ] hence their values are fully determined by those of the eigenvalue @xmath237 of the hidden global @xmath5 symmetry generator and @xmath13-spin @xmath26 or spin @xmath27 , respectively .
this reveals that @xmath238 and @xmath239 are not independent quantum numbers . the physics behind the hidden @xmath5 symmetry found in ref .
@xcite includes that brought about by the rotated - electron occupancy configurations of the set of @xmath230 $ ] sites of eq .
( [ c - c ] ) .
the use of the corresponding model global @xmath3 symmetry reveals that the numbers @xmath240 are always even integers .
moreover , the application onto @xmath241 states of the off - diagonal generators of the @xmath13-spin @xmath242 or spin @xmath243 algebra provided in eq .
( [ scs ] ) gives zero .
for such states @xmath244 . in turn
, application of these generators onto @xmath245 states flips the @xmath13-spin ( @xmath246 ) or spin ( @xmath247 ) of a deconfined @xmath13-spinon ( @xmath246 ) or deconfined spinon ( @xmath247 ) but leaves invariant the rotated - electron occupancy configurations of the above considered set of @xmath217 sites .
it follows that such set of @xmath238 ( and @xmath239 ) sites refers to @xmath13-spin - singlet ( and spin - singlet ) configurations involving @xmath231 ( and @xmath232 ) @xmath57 @xmath13-spinons ( and @xmath57 spinons ) and an equal number of @xmath58 @xmath13-spinons ( and @xmath58 spinons ) .
it follows from the above discussions that the total numbers of @xmath13-spinons ( @xmath214 ) and spinons ( @xmath215 ) read , @xmath248 = [ m_{\alpha,-1/2 } + m_{\alpha,+1/2 } ] \ , ; \hspace{0.35 cm } m_{\alpha,\pm 1/2 } = [ m^{de}_{\alpha,\pm 1/2 } + m^{co}_{\alpha}/2 ] \ , ; \hspace{0.25 cm } \alpha = \eta \ , , s \
, , \nonumber \\
m_{\eta } & = & n_{a_{\eta}}^d = [ n_a^d - 2s_c ] \ , ; \hspace{0.35 cm } m_{s } = n_{a_{s}}^d = 2s_c \ , .
\label{m - l - c}\end{aligned}\ ] ] the @xmath13-spinon and spinon operator algebra refers to well - defined subspaces spanned by states whose number of each of these basic objects is conserved and given in eqs .
( [ l - l ] ) , ( [ c - c ] ) , and ( [ m - l - c ] ) .
hence in such subspaces the number @xmath30 of rotated - electron singly occupied sites and the numbers @xmath188 and @xmath221 of sites of the @xmath13-spin and spin effective lattices , respectively , are fixed . for hole concentrations @xmath249 and maximum spin density @xmath250 ( or @xmath251 ) reached at some critical magnetic field @xmath252 aligned parallel to the square - lattice plane for @xmath37 and pointing along the chain for @xmath36 the @xmath52 fermion operators are invariant under the electron - rotated - electron unitary transformation .
there is a corresponding fully polarized vacuum @xmath253 that remains invariant under such a transformation .
it reads , @xmath254 here the @xmath13-spin @xmath11 vacuum @xmath255 associated with @xmath188 deconfined @xmath58 @xmath13-spinons , the spin @xmath11 vacuum @xmath256 with @xmath221 deconfined @xmath58 spinons , and the @xmath52 @xmath5 vacuum @xmath257 with @xmath51 @xmath52 fermions remain invariant under the electron - rotated - electron unitary transformation .
the latter vacuum may be expressed as @xmath258 . here
the vacuum @xmath259 corresponds to the electron and rotated - electron vacuum .
its form is that given in eq .
( [ vacuum ] ) with @xmath260 and @xmath261 . for @xmath8 only for a @xmath250
fully polarized state are the occupancy configurations of the state @xmath257 and the corresponding @xmath51 @xmath52 fermions invariant under the electron - rotated - electron unitary transformation .
for the corresponding vacuum @xmath255 ( and @xmath256 ) , the @xmath262 deconfined @xmath58 @xmath13-spinons refer to @xmath263 sites of the original lattice unoccupied by rotated electrons ( and the @xmath264 deconfined @xmath58 spinons to the spins of @xmath265 spin - up rotated electrons that singly occupy sites of such a lattice ) . at maximum spin density
@xmath250 the @xmath52 fermions are the non - interacting spinless fermions that describe the charge degrees of freedom of the electrons of the fully polarized ground state . at that spin density
there are no electron doubly occupied sites and the quantum problem is non - interacting for @xmath8 .
according to the analysis of appendix b , full information about the quantum problem described by the hamiltonian ( [ h ] ) can be achieved by defining it in the lws subspace .
below in section iv we then confirm that within the description introduced in this paper , out of the @xmath266=[2s_{\alpha } + m^{co}_{\alpha}]$ ] sites of the @xmath13-spin ( @xmath214 ) and spin ( @xmath215 ) effective lattice , the @xmath218 sites whose occupancies @xmath13-spin ( @xmath214 ) and spin ( @xmath215 ) degrees of freedom are invariant under electron - rotated - electron unitary transformation as defined in this paper play the role of unoccupied sites . in turn , the remaining @xmath217 sites play the role of occupied sites .
this is a natural consequence of the @xmath13-spin @xmath11 vacuum @xmath255 ( and spin @xmath11 vacuum @xmath256 ) being for all @xmath10 and @xmath43 values invariant under the electron - rotated - electron unitary transformation .
indeed that vacuum is such that @xmath267 ( and @xmath268 ) , so that @xmath269 ( and @xmath270 . )
here we use the 1d problem to illustrate the independence of the @xmath52 effective lattice , @xmath13-spin effective lattice , and spin effective lattice occupancy configurations .
indeed the 1d ba solution is constructed to inherently such configurations being independent .
we then extend our analysis to the model on the square lattice . for simplicity , in the following
we denote the number of sites of the @xmath52 , @xmath13-spin , and spin effective lattices of the 1d model by @xmath191 , @xmath192 , and @xmath193 , respectively , rather than by our general notation @xmath233 , @xmath188 , and @xmath221 with @xmath36 for the 1d lattice .
( the @xmath52 effective lattice is identical to the original lattice . ) as discussed above and confirmed below in section iv - d and in ref .
@xcite , the representations of the model global @xmath6/z_2 ^ 2 $ ] symmetry are naturally expressed in terms of rotated - electron occupancy configurations rather than of electron configurations .
hence a quantity of interest for our analysis is the amplitude associated with in a given energy eigenstate @xmath272 , @xmath271 rotated electrons of spin projection @xmath273 , rather than the corresponding @xmath271 electrons , being found at positions @xmath274 , respectively , @xmath275 here the @xmath276 electron and rotated - electron vacuum @xmath277 refers to the vacuum @xmath278 of eq .
( [ vacuum ] ) for @xmath279 , @xmath280 , and @xmath281 and @xmath272 is an arbitrary energy eigenstate . obviously , the state @xmath282 involved in the amplitude ( [ amplitude-1d - rot ] ) is not an energy eigenstate .
application onto @xmath272 of the product operator @xmath283 provides important information . as justified below , before removing the rotated electrons and corresponding electrons such an operator breaks the @xmath231 @xmath13-spin - neutral pairs involving @xmath231 confined @xmath57 @xmath13-spinons and @xmath231 confined @xmath58 @xmath13-spinons as well as the @xmath232 spin - neutral pairs involving @xmath232 confined @xmath57 spinons and @xmath232 confined @xmath58 spinons . for the electron - rotated - electron unitary operator @xmath106 as defined in section ii one can profit both from @xmath101 being an energy eigenstate for @xmath8 and the electronic vacuum @xmath277 associated with an empty original lattice being invariant under @xmath106 , so that @xmath284 .
combining these two properties with the unitarity of @xmath106 one may express the amplitude ( [ amplitude-1d - rot ] ) for the @xmath271 rotated electrons at finite @xmath10 in terms of the corresponding amplitude of @xmath271 electrons for @xmath48 , @xmath285 however we recall that in the @xmath48 limit electrons and rotated electrons become the same objects so that the amplitude @xmath286 also refers to @xmath271 rotated electrons . hence it is the amplitude @xmath287 of eq .
( [ amplitude-1d - rot ] ) for @xmath48 .
this reveals that for @xmath8 and the rotated electrons generated from the electrons by the electron - rotated - electron unitary operator @xmath106 as defined in section ii the amplitude ( [ amplitude-1d - rot ] ) is independent of @xmath10 .
fortunately , the amplitude for @xmath271 electrons can be extracted from the exact ba solution .
for @xmath10 finite it involves a permutation that is extremely complicated and useless for practical calculations @xcite . in turn , in the @xmath48 limit of eq .
( [ amplitude-1d - rot - elec ] ) it has a much simpler form first given in eq .
( 2.23 ) of ref .
@xcite for energy eingenstates that are lwss of both the @xmath13-spin and spin algebras .
it reads , @xmath288 = ( -1)^q\,\det\left[e^{i k_{pj}x^s_{qj}}\right ] \
, , \nonumber \\
\phi^{s_{\eta}}_{su(2 ) } ( x_1^d , x_2^d, ...
,x_{m^{co}_{\eta}/2}^d ) & = & e^{i\pi m^{co}_{\eta}/2}\,\phi_1 ( x_1^d , x_2^d, ... ,x_{m^{co}_{\eta}/2}^d ) \ , , \nonumber \\ \phi^{s_{s}}_{su(2 ) } ( x_1^{s\downarrow},x_2^{s\downarrow}, ... ,x_{m^{co}_{s}/2}^{s\downarrow } ) & = & \phi_2 ( x_1^{s\downarrow},x_2^{s\downarrow}, ... ,x_{m^{co}_{s}/2}^{s\downarrow } ) \ , . \label{amplitude-1d}\end{aligned}\ ] ] here we used the notations of our general description , which are those suitable to the model global @xmath6/z_2 ^ 2 $ ] symmetry , the permutation @xmath289 is defined by the condition , @xmath290 the summation @xmath291 is extended over all permutations @xmath292 of the ba real rapidity momenta @xmath293 and the corresponding determinant depends only on the spatial - coordinates of the @xmath30 sites singly occupied by the rotated electrons of eq .
( [ q ] ) , and not on their spins @xmath294 .
the functions @xmath295 and @xmath296 are obtained on taking the @xmath48 limit of the expressions defined in eq .
( 2.24 ) of ref .
@xcite and in the paragraph after that equation .
they are such that the corresponding functions @xmath297 and @xmath298 have the following limiting behavior @xmath299 = [ m_{\alpha}/2-s_{\alpha } ] = 0 \ , , \hspace{0.35 cm } \alpha = \eta , s \ , .
\label{calpha-0}\ ] ] the ba solution refers either to a lws or a hws of both the @xmath13-spin and spin algebras .
expression ( [ amplitude-1d ] ) refers to a lws for which the sites occupied by the @xmath300=[2s_s + m^{co}_s+m^{co}_{\eta}]$ ] rotated electrons include the @xmath301 $ ] singly occupied sites whose spatial coordinates @xmath302 are those in the argument of the function @xmath303 plus the @xmath231 doubly occupied sites whose spatial coordinates @xmath304 are those in the argument of the function @xmath305 .
moreover , the spatial coordinates of the @xmath306 $ ] rotated - electron unoccupied sites are those left over by the above @xmath307=[2s_s + m^{co}_s+m^{co}_{\eta}/2]$ ] sites . in turn , the sites occupied by the @xmath308 $ ] spin - down rotated electrons include @xmath309 singly occupied sites whose spatial coordinates @xmath310 are those in the argument of the function @xmath311 plus @xmath231 doubly occupied sites whose spatial coordinates are again those in the argument of the function @xmath305 . on the other hand , the sites occupied by the @xmath312 $ ] spin - up rotated electrons include @xmath313 $ ] singly occupied sites plus @xmath231 doubly occupied sites . while the spatial coordinates of the latter @xmath231 doubly occupied sites are the same as those of the corresponding spin - down rotated electrons , the spatial coordinates of the @xmath313 $ ] sites singly occupied by spin - up rotated electrons are those of the @xmath301 $ ] singly occupied sites in the argument of the function @xmath303 except for the @xmath309 sites whose spatial coordinates @xmath314 are those of the singly occupied sites in the argument of the function @xmath311 , which refer only to spin - down rotated electrons .
although the ba solution refers only to lwss or hwss , except for the @xmath315 phase factor in the @xmath305 expression of eq .
( [ amplitude-1d ] ) , which in the general case is replaced by @xmath316 , the amplitude @xmath287 expression given in that equation applies to any energy eigenstate .
furthermore , in the general case the variables of the function @xmath305 are as given in eq .
( [ amplitude-1d ] ) for @xmath145 but change to @xmath317 for @xmath318 .
here the spatial coordinates @xmath319 refer to rotated - electron unoccupied sites associated with occupancy configurations whose @xmath13-spin degrees of freedom are not invariant under the electron - rotated - electron unitary transformation .
also the function @xmath311 remains the same for @xmath320 and is replaced by @xmath321 for @xmath322 .
the spatial coordinates @xmath323 in the argument of the latter function refer to spin - up rotated - electron singly occupied sites associated with occupancy configurations whose spin degrees of freedom are not invariant under the electron - rotated - electron unitary transformation .
we emphasize that the @xmath13-spin function @xmath324 and spin function @xmath325 do not depend on the spatial coordinates of the @xmath208 deconfined @xmath13-spinons and @xmath209 deconfined spinons , respectively . indeed , for @xmath145 and @xmath318 the @xmath13-spin function @xmath324 only depends on the spatial coordinates of the @xmath231 confined @xmath57 @xmath13-spinons and @xmath231 confined @xmath58 @xmath13-spinons , respectively , whose @xmath13-spin - neutral configurations are broken upon the application of the operator @xmath283 onto the energy eigenstate @xmath272 .
similarly , for @xmath320 and @xmath322 the spin function @xmath325 only depends on the spatial coordinates of the @xmath232 confined @xmath57 spinons and @xmath232 confined @xmath58 spinons , respectively , whose spin - neutral configurations are also broken upon the application of that operator onto the energy eigenstate under consideration . importantly and as given in eq .
( [ amplitude-1d ] ) for the lwss of both algebras , for all @xmath326 energy eigenstates the @xmath271 rotated - electron amplitude @xmath287 is a product of three functions @xmath327 , @xmath324 , and @xmath325 associated with the @xmath52 hidden @xmath5 symmetry , @xmath13-spin @xmath11 symmetry , and spin @xmath11 symmetry , respectively , contained in the model global @xmath6/z_2 ^ 2 $ ] symmetry .
furthermore , for @xmath8 that amplitude is independent of @xmath10 . however , the energy of the corresponding occupancy configurations strongly depends on @xmath10 .
this follows from the hamiltonian not commuting with the electron - rotated - electron unitary operator @xmath106 . for @xmath145 ( and @xmath318 )
excited states of physical interest have vanishing rotated - electron ( and rotated - hole ) double occupancy .
our discussion of the @xmath271 rotated - electron amplitude @xmath287 is limited to such states yet , except for the above reported small changes its expression ( [ amplitude-1d - rot ] ) , applies to all @xmath326 energy eingenstates . in turn , at @xmath146 our study refers to both states with vanishing and finite rotated - electron and rotated - hole double occupancy . for the processes discussed below
the large-@xmath10 energy scale @xmath1 is for finite @xmath10 values and @xmath328 replaced by twice the absolute value of chemical potential , @xmath329 .
this is because for @xmath330 ( and @xmath331 ) and finite @xmath10 values the minimum energy for creation onto the ground state of a rotated - electron doubly occupied site ( and unoccupied site ) is given by @xmath329 . in turn , at @xmath146 the chemical potential @xmath332 belongs to the range @xmath333 whose energy width @xmath334 equals the mott - hubbard gap .
we use the convention that for @xmath328 the chemical potential @xmath332 sign is that of the hole concentration @xmath335 . alike for 1d , we consider that for the square lattice the magnetic field points in directions such that @xmath336 for spin density @xmath320 and @xmath337 for @xmath322 .
the minimum magnitude of the energy @xmath338 for creation of a number @xmath339 of rotated - electron doubly occupied sites onto a @xmath12 and @xmath146 ground state and a @xmath12 and @xmath145 ground state is , @xmath340 \ , , \hspace{0.2cm}{\rm at}\hspace{0.2cm}x=0\hspace{0.2cm}{\rm and}\hspace{0.2cm}\mu \in ( -\mu^0,\mu^0 ) \ , , \nonumber \\
& = & 2\mu\,[m^{de}_{\eta,-1/2}+m^{co}_{\eta}/2 ] \ , , \hspace{0.2cm}{\rm for}\hspace{0.2cm}x>0 \ , , \label{min - d}\end{aligned}\ ] ] respectively .
similar expressions apply to the minimum magnitude of the energy @xmath341 for creation of a number @xmath342 of rotated - electron unoccupied sites onto a @xmath12 and @xmath146 ground state and a @xmath12 and @xmath318 ground state , respectively , provided that @xmath332 is replaced by @xmath343 and @xmath224 by @xmath225 . hence for @xmath333 the minimum energy for creation onto the @xmath12 and @xmath146 ground state of one rotated - electron doubly occupied site and rotated - electron unoccupied site is @xmath344 and @xmath345 , respectively . for the chemical potential in the middle of the mott - hubbard gap so that @xmath346 at @xmath146 the minimum energy of either process is given by @xmath347 . for the model on the 1d ( and square ) lattice at @xmath12 the important @xmath146 energy scale @xmath348 appearing in eq .
( [ min - d ] ) , which is the mott - hubbard gap , has the following exact ( and approximate ) limiting behaviors , @xmath349 ^ 2 t\over u}\right)^{d/2}e^{-2\pi \left({t\over u}\right)^{1/d } } \ , , \hspace{0.25 cm } u/4t\ll 1 \ , ; \hspace{0.5 cm } 2\mu^0 \approx [ u - 4dt ] \
, , \hspace{0.25 cm } u/4t\gg 1 \ , , \hspace{0.25 cm } d = 1,2\ , . \label{dmh}\ ] ] these half - filling results are consistent with the properties of the @xmath146 and @xmath12 absolute ground state @xcite . in turn , for @xmath350 and @xmath12 the chemical potential reads @xmath351 for @xmath48 and for finite @xmath10 values is an increasing function of the hole concentration @xmath194 such that , @xmath352 here @xmath353 .
the related energy scale @xmath354 reads , @xmath355 expression ( [ mu-1 ] ) is exact both for the hubbard model on a 1d and square lattice .
it can be explicitly derived for both lattices .
it refers to the non - interacting limit of vanishing electronic density . for the model on the square lattice
the limiting behaviors reported in eq .
( [ dmh ] ) read @xmath356 and @xmath357 for @xmath358 and @xmath206 , respectively .
they are those of the related zero - temperature gap of eq .
( 13 ) of ref .
@xcite , which although showing up in the spin degrees of freedom equals half the charge mott - hubbard gap @xcite .
as discussed below in section iv - c , @xmath145 ( and @xmath318 ) ground states have vanishing rotated - electron ( and rotated - hole ) double occupancy , the same applying for @xmath145 ( and @xmath318 ) to the states of excitation energy @xmath359 that span the one- and two - electron subspace as defined in section v. to understand the 1d amplitude expression ( [ amplitude-1d ] ) , we consider a configuration in which the first @xmath360=n_c$ ] sites in the original lattice are singly occupied and the remaining @xmath361=[2s_{\eta}+m^{co}_{\eta}]=n_c^h$ ] sites are unoccupied or doubly occupied by rotated electrons . in this configuration and
considering processes of energy lower than @xmath329 the rotated electrons can not move ( except for the last one ) . indeed either the pauli principle or the onsite repulsion prevents it .
although in this configuration there is no direct interaction between the rotated electrons , through an intermediate state with energy @xmath329 neighboring rotated electrons can see each other s spins , and rotated electrons with different spins can change position .
we recall that for @xmath362 the energy scale @xmath329 reads @xmath1 for all values of @xmath194 and the usual @xmath206 discussion in terms of electrons is recovered @xcite .
for instance , it follows from the above analysis that for @xmath206 the spins can move in the same way as do those in a heisenberg chain . for @xmath206
the distribution of the spins will correspond to the eigenstates of the heisenberg hamiltonian .
the situation with the rotated - electron unoccupied and doubly occupied sites is similar .
neighboring sites can observe each others occupancy through an intermediate state of relative energy whose absolute value is @xmath329 .
moreover , the same intermediate state makes it possible for a rotated - electron unoccupied and doubly occupied site to interchange position .
( thus for @xmath206 the distribution of the unoccupied and doubly occupied sites will be the same as the distribution of up and down spins in a heisenberg chain . )
the main point here is that neither the spinon distribution of the function @xmath363 where @xmath72 and @xmath73 for @xmath320 and @xmath322 , respectively , nor the @xmath13-spinon distribution associated with the relative distribution of the rotated - electron unoccupied and doubly occupied sites of the function @xmath305 for @xmath145 and @xmath317 for @xmath318 do change if the chain of rotated - electron singly occupied sites is `` diluted '' by rotated - electron unoccupied and doubly occupied sites , making possible also direct propagation for the rotated electrons .
such a propagation is in turn described by independent occupancy configurations associated with the @xmath52 fermion distribution of the function @xmath303 .
this important property is behind the independence of the @xmath52 effective lattice , @xmath13-spin effective lattice , and spin effective lattice occupancy configurations . indeed , within our description the three degrees of freedom associated with the three types of rotated - electron occupancy configurations behind the functions @xmath303 , @xmath305 , and @xmath311 in the amplitude general expression of eq .
( [ amplitude-1d ] ) are naturally described in terms of independent @xmath52 fermion , @xmath13-spinon , and spinon occupancy configurations , respectively .
the relationship of the operators of the @xmath52 fermions to those of the rotated electrons is provided in eq .
( [ fc+ ] ) . that of the operators of the @xmath13-spinons and spinons to the rotated - electron operators is given in eqs .
( [ sir - pir])-([rotated - quasi - spin ] ) . except for @xmath206 , the above independence of the three types of occupancy configurations
is lost if one uses descriptions based on electrons rather than rotated electrons as defined in section ii . indeed and except for @xmath206 , for finite @xmath10 values the three independent degrees of freedom associated with the @xmath52 hidden @xmath5 symmetry , @xmath13-spin @xmath11 symmetry , and spin @xmath11 symmetry , respectively , contained in the model global @xmath6/z_2 ^ 2 $ ] symmetry , are difficult to separate in terms of electron occupancy configurations .
unlike in 1d , for the model on the square lattice the @xmath271 rotated - electron amplitudes can not be expressed in terms of simple permutations and site occupancy chain orders .
nonetheless , due to the common interplay between the on - site electron correlations and the global @xmath6/z_2 ^ 2 $ ] symmetry , some of the physics is the same , particularly in the @xmath196 limit .
one may consider a configuration in which a original - lattice compact square domain of @xmath364=n_c$ ] sites whose edge contains @xmath365 sites is singly occupied by rotated electrons , and the complementary two - dimensional ( 2d ) compact domain of @xmath366= [ 2s_{\eta}+m^{co}_{\eta}]=n_c^h$ ] sites refers to rotated - electron unoccupied or doubly occupied sites .
obviously , in 2d there are many other shapes for compact domains of @xmath366= [ 2s_{\eta}+m^{co}_{\eta}]=n_c^h$ ] sites
. the square shape will be justified below . in this configuration and considering processes of energy lower than @xmath329 , either the pauli principle or the onsite repulsion prevents the rotated electrons to move ( except for those on the lines referring to the square edges separating the two 2d domains ) .
although in this configuration there is no direct interaction between the rotated electrons , through an intermediate state with energy @xmath329 neighboring rotated electrons can see each other s spins , and rotated electrons with different spins can change position . alike for 1d , the situation with the rotated - electron unoccupied and doubly occupied sites is similar for a original - lattice compact square domain of @xmath366=[2s_{\eta}+m^{co}_{\eta}]=n_c^h$ ] sites . unfortunately and unlike for 1d , for a finite system the corresponding spinon and @xmath13-spinon distributions may change if the the square - shape compact domain of rotated - electron singly occupied sites is `` diluted '' by rotated - electron unoccupied and doubly occupied sites .
indeed , for a 2d system there is no order equivalent to the 1d uniquely defined chain order .
however , within the @xmath367 limit that the present description refers to the _ dominant _ @xmath52 fermion occupancy configurations refer to an average uniform distribution of the @xmath52 effective lattice occupied sites . specifically , the @xmath52 fermion positions of such a uniform configuration correspond to the average positions of the @xmath52 fermions in an energy eigenstate
indeed , the @xmath52 fermion momentum occupancy configurations of such a state are a superposition of all compatible real - space @xmath52 effective lattice occupancy configurations . fortunately , for the average configuration in which the rotated - electron singly occupied sites are uniformly `` diluted '' by rotated - electron unoccupied and doubly occupied sites the spinon and @xmath13-spinon distributions of the above original - lattice square - shape compact domain of @xmath364=n_c$ ] sites do not change
. they may be described by corresponding occupancy configurations of an effective square lattice with @xmath364=n_c$ ] sites and edge length @xmath368 whose average spacing is for @xmath328 larger than that of the original lattice .
the same arguments apply to the @xmath13-spinon distributions .
this reveals that for the hubbard model on the square lattice with @xmath367 sites the concepts of a spin effective square lattice and @xmath13-spin square effective lattice apply provided that @xmath369 and @xmath194 are finite , respectively . that the above considered initial reference original - lattice compact domain of @xmath364=n_c$ ] ( and @xmath366=[2s_{\eta}+m^{co}_{\eta}]=n_c^h$ ] ) sites has a square shape follows from only that compact domain shape transforming into a corresponding uniform `` diluted '' domain fully contained in the original square lattice .
any other initial reference compact - domain shape would transform into a uniform `` diluted '' domain that is not fully contained in the original square lattice and thus is neither physically nor mathematically acceptable .
here we are considering that the 2d crystal has a square shape whose edge is @xmath370 . if the square - lattice crystal has any other shape the compact domain should have a similar shape with an area reduced by a factor of @xmath371 for the spin effective lattice ( and of @xmath372/n_a^2 $ ] for the @xmath13-spin effective lattice ) .
only then it transforms into a corresponding uniform `` diluted '' domain fully contained in the crystal .
the final result is though the same , the obtained effective lattice being a square lattice whose spacing is given by , @xmath373 where @xmath374 .
this spinon and @xmath13-spinon distributions `` average invariance '' under `` site dilution '' emerging for the square lattice in the @xmath196 limit is behind the description of such distributions in terms of the occupancy configurations of independent @xmath13-spin and spin effective square lattices , respectively , and the direct propagation of @xmath52 fermions in terms of occupancy configurations in an independent @xmath52 effective lattice .
while for @xmath350 and @xmath375 the former two lattices have both for 1d and 2d a number of sites smaller than @xmath376 , the @xmath52 effective lattice is identical to the original lattice . within the @xmath33 limit of our description the concept of a spin ( and @xmath13-spin ) effective lattice
is well defined for finite values of the electronic density @xmath369 ( and hole concentration @xmath194 ) .
the reasoning for the validity of the use of the corresponding effective lattices occupancy configurations may be summarized by the two following statements : \1 ) the representation associated with the present description contains full information about the relative positions of the sites of the @xmath13-spin and spin effective lattices in the original lattice . for each energy - eigenstate
rotated - electron real - space occupancy configuration , that information is stored in the corresponding occupancy configurations of the @xmath52 fermions in their @xmath52 effective lattice .
the latter lattice is identical to the original lattice .
such configurations correspond to the state representations of the @xmath5 symmetry in the subspaces spanned by states with fixed values of @xmath28 , @xmath26 , and @xmath27 .
indeed , the sites of the @xmath13-spin ( and spin ) effective lattice have in the original lattice the same real - space coordinates as the sites of the @xmath52 effective lattice unoccupied ( and occupied ) by @xmath52 fermions .
\2 ) within the @xmath33 limit that our description refers to , provided that the electronic density @xmath369 ( and hole concentration @xmath194 ) is finite , the dominant @xmath52 effective lattice occupancy configurations of an energy eigenstate of the hubbard model on the square lattice refer to a nearly uniform distribution of the @xmath52 fermions occupied sites ( and unoccupied sites ) .
hence due to the spinon and @xmath13 spinon distribution `` average order '' emerging for the model on the square lattice in the @xmath196 limit , the spin and @xmath13-spin effective lattices may be represented by square lattices .
moreover , the chain order invariance occurring for the 1d model both for the finite system and in that limit justifies why such effective lattices are 1d lattices . for both models the corresponding spin effective lattice spacing @xmath377 and @xmath13-spin effective lattice spacing @xmath378 refers to the average spacing between the @xmath52 effective lattice occupied sites and between such a lattice unoccupied sites , respectively , given in eq .
( [ a - alpha ] ) . such spin and @xmath13-spin effective lattices obey the physical requirement condition that in the @xmath379 and @xmath380 limit , respectively , equal the original lattice .
note that in the @xmath379 ( and @xmath380 ) limit one has that @xmath381 and the @xmath13-spin effective lattice does not exist ( and @xmath260 and the spin effective lattice does not exist . )
the validity for @xmath33 of the concept of a spin effective square lattice as constructed in this paper is confirmed by the behavior of the expectation value @xmath382 of any energy eigenstate @xmath383 . here
@xmath384 is the operator associated with the distance in real space of any of the @xmath221 sites of the spin effective lattice from the rotated - electron singly occupied site of the original lattice closest to it .
the state @xmath383 belongs to a subspace with fixed number of rotated - electron singly occupied sites .
the point is that @xmath382 vanishes in the @xmath196 limit .
we recall that the @xmath198 sites of the spin effective lattice occupancies describe the spin degrees of freedom of the @xmath51 rotated - electron singly occupied sites of the original lattice .
the same applies to the @xmath13-spin effective lattice . consistently with the expression @xmath385\,a$ ] of eq .
( [ a - alpha ] ) where @xmath386 , the @xmath13-spin ( and spin ) effective lattice has both for 1d and 2d the same length and edge length @xmath368 , respectively , as the original lattice .
furthermore , the requirement that for the 2d case when going through the whole crystal of square shape along the @xmath387 or @xmath388 directions a @xmath13-spinon ( and spinon ) passes an overall distance @xmath368 is met by an effective @xmath13-spin ( and spin ) square lattice .
since the number of sites sum - rule @xmath389=n_a^d$ ] holds , the @xmath13-spin and spin effective lattices have in general a number of sites @xmath188 and @xmath221 , respectively , smaller than that of the original lattice , @xmath233 .
it follows that their lattice spacings ( [ a - alpha ] ) are larger than that of the original lattice .
the transformation laws under the electron - rotated - electron unitary transformation of the @xmath13-spin and spin degrees of freedom of the rotated - electron occupancy configurations that generate the energy eigenstates are in 1d behind the @xmath13-spin function @xmath324 and spin function @xmath325 of the @xmath271 rotated - electron amplitude expression ( [ amplitude-1d ] ) not depending on the spatial coordinates of the deconfined @xmath13-spinons of @xmath13-spin projections @xmath57 and @xmath58 and deconfined spinons of spin projections @xmath57 and @xmath58 , respectively . both for the model on the 1d and square lattice
, it follows from such transformation laws that the @xmath13-spin and spin degrees of freedom of the rotated - electron occupancy configurations that generate an energy eigenstate may be described in terms of @xmath390 ( and @xmath391 ) unoccupied sites " and the @xmath238 ( and @xmath239 ) occupied sites " of the @xmath13-spin ( and spin ) effective lattice .
the corresponding @xmath13-spinon and spinon occupancy configurations generate state representations of the model global @xmath3 symmetry .
the @xmath392 ( and @xmath393 ) unoccupied sites " of such a @xmath13-spin ( and spin ) effective lattice correspond to @xmath394 deconfined @xmath13-spinons ( and @xmath395 deconfined spinons ) that remain invariant under the electron - rotated - electron unitary transformation . in turn , the @xmath231 @xmath58 @xmath13-spinons and @xmath231 @xmath57 @xmath13-spinons ( and @xmath232 @xmath58 spinons and @xmath232 @xmath57 spinons ) that refer to the @xmath238 ( and @xmath239 ) occupied sites " of such a lattice do not remain invariant under that unitary transformation . for the hubbard model on the square lattice
the above concepts of @xmath13-spin and spin effective lattices occupied sites " and unoccupied sites " apply both to the energy eigenstates and the related complete set of states introduced below in section iv - e . for @xmath145
the subspace of more physical interest is that of vanishing rotated - electron double occupancy . for simplicity
we consider here its lws subspace , for which @xmath396 in eq .
( [ l - l ] ) and @xmath269 in eq .
( [ 2s-2c ] ) with @xmath214 in both these equations . since @xmath397=[n_a^d-2s_c-2s_{\eta}]=0 $ ] and thus @xmath398 , for the model in that subspace the number of sites of the spin effective lattice @xmath399 of eq .
( [ na - eta - s ] ) and its spacing given in eq .
( [ a - alpha ] ) for @xmath215 simplify to , @xmath400 respectively .
such a @xmath401 subspace is spanned by ground states and their excited states of energy below @xmath402 . for the 1d case , combination of the general expression ( [ amplitude-1d ] ) with the boundary condition @xmath403 of eq .
( [ calpha-0 ] ) immediately leads for such a @xmath145 and @xmath401 subspace to the expression ( 2.14 ) of ref .
@xcite , which refers to the @xmath271 electrons amplitude for @xmath48 .
we recall that here it refers as well to the @xmath271 rotated - electrons amplitude for @xmath8 .
the sites of the @xmath13-spin lattice refer to those doubly occupied and unoccupied by rotated electrons in the original lattice .
however , for the hubbard model in the vanishing rotated - electron double occupancy subspace the concept of a @xmath13-spin lattice is useless . indeed , for that subspace such a lattice either is empty ( @xmath145 ) or does not exist ( @xmath146 ) .
this is because the @xmath13-spin degrees of freedom of the states that span that subspace are the same as those of the @xmath404 vacuum @xmath255 of eq .
( [ vacuum ] ) . for states
for which @xmath405 and thus @xmath406 the @xmath13-spin lattice does not exist and hence the spin effective lattice is identical to the original lattice . in turn , for @xmath407 lwss the @xmath13-spin degrees of freedom correspond to a single occupancy configuration of the @xmath263 deconfined @xmath58 @xmath13-spinons . only for states and subspaces
for which @xmath408 $ ] is finite and the inequality @xmath409 holds is the concept of a @xmath13-spin effective lattice useful .
for @xmath145 the subspace resulting from the overlap of the @xmath401 subspace with the one- and two - electron subspace defined below in section v plays a major role in the one- and two - electron physics .
the goal of this section is the introduction of a useful complete set of state representations of the model global @xmath3 symmetry .
such states are constructed to inherently within the @xmath196 limit being momentum eigenstates . besides a momentum eigenvalue
, such states have fixed values of @xmath26 , @xmath199 , @xmath27 , @xmath200 , and @xmath28 .
the interest of such states is that for hubbard model on the square lattice in the one- and two - electron subspace introduced in section v , which refers to the square - lattice quantum liquid of ref .
@xcite , they are both energy and momentum eigenstates . since they refer to a complete set of states , the general energy and momentum eigenstates @xmath101 that span the whole hilbert space of that model are a superposition of a well - defined sub - set of such states with the same momentum eigenvalue and the same @xmath26 , @xmath199 , @xmath27 , @xmath200 , and @xmath28 values and thus @xmath410 $ ] and @xmath411 $ ] values .
as discussed in appendix a , for the 1d hubbard model , due to the occurrence of an infinite number of conservation laws , such states are both energy and momentum eigenstates , so that the above superpositions refer to a single state .
a preliminary version of the general rotated - electron operator description introduced here , which lacked its relation to the model global @xmath3 symmetry , was presented in ref .
@xcite for the particular case of the 1d hubbard model . on expressing the rotated - electron quantum numbers in terms of those of the exact ba solution ,
a description in terms of @xmath52 fermions without internal structure and several several branches of @xmath13-spin - singlet @xmath59-@xmath13-spinon composite objects called in this paper @xmath60 fermions and several branches of spin - singlet @xmath59-spinon composite objects called in it @xmath61 fermions emerges . here
@xmath62 is the number of @xmath13-spinon and spinon pairs .
in addition , the above states have a well - defined number of deconfined @xmath13-spinons and deconfined spinons . inspired in the 1d exact solution , in the following we construct similar states for the model on the square lattice .
the complexity of the expressions of the @xmath60 and @xmath61 fermion operators in terms of @xmath13-spinon and spinon operators , respectively , increases upon increasing the number @xmath59 of confined @xmath13-spinons and confined spinons .
the expressions of the spin - neutral @xmath59-spinon @xmath61 fermion operators involve products of the spinon operators @xmath163 . similarly , those of the @xmath13-spin - neutral @xmath59-@xmath13-spinon @xmath60 fermion operators involve products of the @xmath13-spinon operators @xmath164 .
the three spinon operators @xmath163 and three @xmath13-spinon operators @xmath164 are expressed in terms of the rotated - electron operators in eqs .
( [ sir - pir])-([rotated - quasi - spin]).the simplest general expressions of spin - neutral two - spinon @xmath64 fermion operators in terms of the spinon operators @xmath163 are given in section vi both for the model on the 1d and square lattices .
the global @xmath3 symmetry of the model on any bipartite lattice implies that some features of the corresponding state representations are common to all such lattices .
our extension of the 1d model @xmath411$]-site spin - neutral and @xmath410$]-site @xmath13-spin - neutral rotated - electron occupancy configurations in terms of those of composite @xmath61 fermions and @xmath60 fermions , respectively , to the model on the square lattice accounts for the basic differences between the physics of the two models . within the rotated - electron occupancy configurations that generate the exact energy eigenstates
@xmath101 considered in section ii , there are @xmath230 $ ] sites out of the @xmath233 sites of the original lattice whose rotated - electron occupancy configurations are not invariant under the electron - rotated - electron unitary transformation . the @xmath13-spin ( and spin )
degrees of freedom of @xmath238 ( and @xmath239 ) of such sites refer to @xmath13-spin - neutral ( and spin - neutral ) configurations involving @xmath231 @xmath58 @xmath13-spinons ( and @xmath232 @xmath58 spinons ) and an equal number @xmath231 of @xmath57 @xmath13-spinons ( and @xmath232 of @xmath57 spinons ) .
this holds as well for the related momentum eigenstates @xmath105 considered in the following . for the model on the square lattice ,
the latter states refer in general to a partition different from that of the energy eigenstates @xmath101 of the @xmath13-spin - neutral ( and spin - neutral ) configurations of such @xmath238 ( and @xmath239 ) @xmath13-spinons ( and spinons ) , in terms of smaller configurations .
specifically , for a given momentum eigenstate there is for each branch involving @xmath62 pairs of @xmath13-spinons ( and spinons ) a well - defined number @xmath412 of @xmath13-spin - neutral @xmath59-@xmath13-spinon composite @xmath60 fermions ( and @xmath413 of spin - neutral @xmath59-spinon composite @xmath61 fermions ) .
the set of such composite objects describes the @xmath13-spinon ( and spinon ) occupancy configurations of exactly @xmath238 ( and @xmath239 ) sites of the original lattice .
hence the following two sum rules hold for all momentum eigenstates , @xmath414 = 2\sum_{\nu = 1}^{\infty}\nu\,n_{\alpha\nu } \ , ; \hspace{0.35 cm } \alpha = \eta \ , , s \ , .
\label{m - l - sum}\ ] ] such sum rules refer to subspaces spanned by states with fixed values of @xmath28 , @xmath26 , and @xmath27 . one @xmath13-spin neutral @xmath59-@xmath13-spinon composite @xmath60 fermion describes the @xmath13-spin degrees of freedom of a @xmath13-spin - singlet occupancy configuration involving @xmath415 sites of the original lattice unoccupied by rotated electrons and an equal number of sites doubly occupied by rotated electrons .
the remaining degrees of freedom of such a rotated - electron occupancy configuration are described by @xmath59 unoccupied sites of the @xmath52 effective lattice whose spatial coordinates are those of the corresponding @xmath59 sites of the original lattice .
similarly , one spin neutral @xmath59-spinon composite @xmath61 fermion describes the spin degrees of freedom of a spin - singlet occupancy configuration involving @xmath416 sites of the original lattice singly occupied by spin - up rotated electrons and an equal number of sites singly occupied by spin - down rotated electrons .
the remaining degrees of freedom of that rotated - electron occupancy configuration are described by @xmath59 occupied sites of the @xmath52 effective lattice whose spatial coordinates are those of the corresponding @xmath59 sites of the original lattice . for each @xmath60 fermion branch ( and @xmath61 fermion branch ) , one may consider a @xmath60 effective lattice ( and @xmath61 effective lattice ) .
it refers to occupancy configurations of @xmath13-spin - neutral ( and spin - neutral ) bonds of @xmath59 confined @xmath13-spinons ( and spinons ) .
hence each `` occupied site '' of such an effective lattice corresponds to @xmath59 sites of the @xmath13-spin ( and spin ) effective lattice . in turn , the @xmath394 ( and @xmath417 ) @xmath13-spin ( and spin ) effective lattice sites referring to the deconfined @xmath13-spinons ( and deconfined spinons ) and some of such a lattice sites referring to @xmath418-@xmath13-spinon composite @xmath419 fermions ( and @xmath418-spinon composite @xmath420 fermions ) of @xmath421 branches are found below to play the role of `` unoccupied sites '' of such a @xmath60 effective lattice ( and @xmath61 effective lattice ) .
the conjugate of the @xmath63 effective lattice site space variables are the @xmath63 band discrete momentum values .
their number equals that of the @xmath63 effective lattice sites .
( for 1d , such a momentum values are good quantum numbers . )
as confirmed below in section iv - d , the states generated by @xmath52 fermion and @xmath63 fermion occupancy configurations in the corresponding @xmath52 and @xmath63 momentum bands , respectively , are state representations of the global @xmath3 symmetry . within chromodynamics
the quarks have color but all quark - composite physical particles are color - neutral @xcite . here the @xmath13-spinon ( and spinons ) that are not invariant under the electron - rotated - electron unitary transformation have @xmath13-spin @xmath9 ( and spin @xmath9 ) but are confined within @xmath13-spin - neutral ( and spin - neutral ) @xmath59-@xmath13-spinon ( and @xmath59-spinon ) composite @xmath60 fermions ( and @xmath61 fermions ) . the exact and detailed internal @xmath59-spinon configuration and @xmath59-@xmath13-spinon configuration of a composite @xmath61 fermion and @xmath60 fermion , respectively ,
is in general an involved unsolved problem .
( in 1d the exact ba solution takes implicitly into account such internal configurations . ) fortunately , however , the problem simplifies for the model in the one- and two - electron subspace introduced in section v for which the only composite object that plays an active role is the two - spinon @xmath64 fermion .
its internal structure is an issue studied in section vi .
such a two - spinon object is related to the resonating - valence - bond pictures considered long ago @xcite for spin - singlet occupancy configurations of ground states . in order to have control over the number of state representations of the model global @xmath3 symmetry , in the following we consider all multi-@xmath13-spinon and multi - spinon composite objects .
full information on the @xmath59-@xmath13-spinon @xmath422 or @xmath59-spinon @xmath423 configurations associated with the internal degrees of freedom of the composite @xmath63 fermions is not needed for the goals of this paper .
indeed , within the present @xmath33 limit the problem of the internal degrees of freedom of the composite @xmath63 fermions and @xmath63 bond particles separates from that of their positions in the corresponding effective lattices .
the partial information on the internal degrees of freedom of the composite @xmath63 fermions needed for our studies is accessed in the following by suitable use of their transformations laws under the electron - rotated - electron unitary transformation . in turn , the deconfined @xmath13-spinons and deconfined spinons are invariant under that transformation .
thus they are non - interacting deconfined objects that are not part of composite @xmath60 fermions and composite @xmath61 fermions , respectively .
strong evidence that the extension from the 1d model to the hubbard model on the square lattice in the one- and two - electron subspace of the @xmath52 fermion and @xmath64 fermion description that results from our general @xmath52 fermion and @xmath63 fermion description is correct provided that the basic differences between the two models are accounted for is given in ref .
@xcite , concerning the half - filled spin spectrum .
( states with finite occupancies of deconfined @xmath13-spinons or one @xmath424 fermion do not contribute to that gapless spectrum . )
the use of the description introduced here reveals that in terms of @xmath64 fermion spinon breaking and deconfined spinon processes the microscopic mechanisms that generate the coherent spectral - weight spin - wave spectrum are very simple .
specifically , the two - spinon @xmath64 fermion description is shown in ref .
@xcite to render a non - perturbative many - electron problem studied in ref .
@xcite by a complex alternative method that involves summation of an infinite set of ladder diagrams into a mere two-@xmath64-fermion - hole spectrum , described by simple analytical expressions .
importantly , for @xmath425 and @xmath426 mev the spin - wave spectrum of the parent compound la@xmath427cuo@xmath428 ( lco ) @xcite is quantitatively described by the corresponding theoretical spectrum derived in ref .
@xcite from simple spinon pair breaking @xmath64 fermion processes .
the lws vacuum ( [ vacuum ] ) of the theory corresponds to @xmath429 deconfined @xmath58 @xmath13-spinons and @xmath430 deconfined @xmath58 spinons .
( the corresponding hws vacuum refers to @xmath431 deconfined @xmath57 @xmath13-spinons and @xmath432 deconfined @xmath57 spinons . )
hence and as mentioned above , such objects play the role of unoccupied sites " of the @xmath13-spin and spin effective lattices , respectively .
consistently , the latter objects have vanishing energy .
relative to that vacuum their @xmath13-spin and spin flip processes correspond to creation " processes of deconfined @xmath57 @xmath13-spinons and deconfined @xmath57 spinons , respectively .
we start by considering processes that conserve the numbers @xmath433 of @xmath13-spinons and @xmath434 of spinons .
such processes also conserve the number of @xmath52 fermions and refer to a subspace whose lws vacuum is that provided in eq .
( [ vacuum ] ) . in turn , creation ( and annihilation ) of a @xmath52 fermion involves annihilation ( and creation ) of a @xmath52 fermion hole .
the latter process involves removal ( and addition ) of one site from ( and to ) the @xmath13-spin effective lattice and addition ( and removal ) of one site to ( and from ) the spin effective lattice .
within our lws representation , creation of a local @xmath60 fermion involves the replacement of @xmath435 deconfined @xmath58 @xmath13-spinons by a suitable @xmath13-spin - neutral configuration involving a number @xmath62 of @xmath57 @xmath13-spinons and an equal number of @xmath58 @xmath13-spinons .
the initial - state @xmath435 deconfined @xmath58 @xmath13-spinons correspond to @xmath435 sites of the @xmath13-spin effective lattice . in turn
, creation of a local @xmath61 fermion involves the replacement of @xmath435 deconfined @xmath58 spinons by a suitable spin - neutral configuration involving a number @xmath62 of @xmath57 spinons and an equal number of @xmath58 spinons .
again the initial - state @xmath435 deconfined @xmath58 spinons refer to @xmath435 sites of the spin effective lattice .
creation of local @xmath60 ( and @xmath61 ) fermions always involves virtual processes where @xmath435 deconfined @xmath58 @xmath13-spinons ( and @xmath435 deconfined @xmath58 spinons ) are replaced by the @xmath13-spin - singlet ( and spin - singlet ) @xmath59-site occupancy configurations of the local @xmath60 fermions ( and @xmath61 fermions ) in the @xmath13-spin ( and spin ) effective lattice . for instance , a given process for which two local @xmath64 fermions of the initial state are replaced by one local @xmath436 fermion in the final state is divided into two virtual processes .
first , two local @xmath64 fermions are annihilated .
this means that the four sites of the spin effective lattice occupied in the initial state by the local @xmath64 fermions are under two spin - flip processes occupied in an intermediate virtual state by four deconfined @xmath58 spinons ( annihilation of two local @xmath64 fermions ) .
second , one local @xmath436 fermion is created on such four sites . that involves two opposite spin - flip processes and rearrangement of the spinons associated with the creation of the local @xmath436 fermion spin - neutral four - spinon occupancy configurations in the spin effective lattice .
the @xmath435 sites of the spin effective lattice occupied by one local @xmath61 fermion correspond to the spin @xmath11 degrees of freedom of @xmath62 spin - up rotated - electron singly occupied sites and @xmath62 spin - down rotated - electron singly occupied sites .
the corresponding hidden @xmath5 symmetry degrees of freedom of such sites rotated - electron occupancies are described by @xmath435 occupied sites of the @xmath52 effective lattice .
therefore , these @xmath435 rotated - electron singly occupied sites are described both by the local @xmath61 fermion and @xmath435 local @xmath52 fermions . in turn
, the @xmath435 sites of the @xmath13-spin effective lattice occupied by one local @xmath60 fermion correspond to the @xmath13-spin @xmath11 degrees of freedom of @xmath62 rotated - electron doubly occupied sites and @xmath62 rotated - electron unoccupied sites .
their hidden @xmath5 symmetry degrees of freedom are described by @xmath435 unoccupied sites of the @xmath52 effective lattice . as a result ,
the @xmath62 rotated - electron doubly occupied sites and @xmath62 rotated - electron unoccupied sites are described both by the local @xmath60 fermion and @xmath435 local @xmath52 fermion holes ( @xmath52 effective lattice unoccupied sites . ) note that any pair of local @xmath63 and @xmath437 fermions always refer to two _ different _ sets of @xmath435 and @xmath438 sites , respectively , of the original lattice .
creation ( and annihilation ) of one local @xmath52 fermion is a process that involves addition ( and removal ) of one site to ( and from ) the spin effective lattice and removal ( and addition ) of one site from ( and to ) the @xmath13-spin effective lattice .
therefore , the spin and @xmath13-spin effective lattices are exotic .
indeed the number of their sites @xmath439 and @xmath197 $ ] , respectively , varies by @xmath159 and @xmath440 upon creation / annihilation of one @xmath52 fermion .
such processes change the eigenvalue @xmath28 of the generator ( [ or - ope ] ) of the hidden global @xmath5 symmetry .
a subspace with fixed @xmath28 value and hence fixed @xmath441 $ ] and @xmath442 values is associated with a well - defined vacuum @xmath253 of form given in eq .
( [ vacuum ] ) .
an excitation involving a change of such values drives the system into a new subspace referring to a different vacuum @xmath253 . in turn ,
@xmath13-spinon and spinon creation and annihilation processes refer to excitations within the same quantum - liquid subspace .
it is associated with a uniquely defined vacuum @xmath253 .
it follows that from the point of view of the @xmath13-spin and spin degrees of freedom , @xmath52 fermion creation and annihilation processes correspond to a change of quantum system .
indeed , the @xmath13-spin and spin lattices and corresponding number of sites change along with the quantum - system vacuum @xmath253 of eq .
( [ vacuum ] ) .
one can then say that for the @xmath13-spinon and spinon representation there is a different quantum system for each eigenvalue @xmath28 of the generator ( [ or - ope ] ) of the hidden global @xmath5 symmetry . in turn , from the point of view of the degrees of freedom associated with the latter symmetry , the model ( [ h ] ) corresponds to a single quantum system .
the local @xmath52 fermions live on a lattice identical to the original lattice whose number of sites @xmath443 $ ] is fixed .
creation ( and annihilation ) of one local @xmath52 fermion involves a virtual process in which the @xmath13-spin effective lattice removed ( and added ) site is occupied in the initial ( and final ) state by a deconfined @xmath58 @xmath13-spinon .
also the site of the spin effective lattice added ( and removed ) by such an elementary process is occupied in the final ( and initial ) state by a deconfined @xmath58 spinon .
if , as occurs for one - electron addition ( and removal ) , creation ( and annihilation ) of a local @xmath52 fermion involves creation ( and annihilation ) of a local @xmath64 fermion , the overall process is divided into two virtual processes .
for instance , complementarily to creation of the local @xmath52 fermion in its effective lattice , the virtual processes occurring in the spin effective lattice are the following .
first , a site occupied by a deconfined @xmath58 spinon is added to that lattice .
second , a local @xmath64 fermion is created .
one of the two initial - state deconfined @xmath58 spinons involved in the final - state two - site @xmath64 bond configuration is that located on the site added to the spin effective lattice .
( a corresponding momentum eigenstate involves the superposition of many local configurations for which that site has different positions . ) on the other hand , if removal of a local @xmath52 fermion from its effective lattice involves annihilation of a local @xmath64 fermion , one has the following virtual processes in the spin effective lattice .
first , a local @xmath64 fermion is annihilated .
this involves a rearrangement process .
it leads to the occupancy of its two sites by two deconfined @xmath58 spinons in the intermediate virtual state .
second , the site of the spin effective lattice occupied by one of these two deconfined spinons is removed along with it . within the lws representation creation ( and annihilation ) of both one local @xmath52 fermion and one local @xmath64 fermion corresponds to creation ( and annihilation ) of a spin - down electron . in turn
, the corresponding process of creation ( and annihilation ) of a spin - up electron involves addition ( and removal ) of one local @xmath52 fermion to ( and from ) its effective lattice and addition ( and removal ) of one site to ( and from ) the spin effective lattice . in the final ( and initial ) state the latter site is occupied by a deconfined @xmath58 spinon .
creation and annihilation of local @xmath52 fermions leads to addition and removal ( and removal and addition ) of sites in the spin ( and @xmath13-spin ) effective lattice , respectively .
similarly , it is confirmed below that creation of one local @xmath444 fermion gives rise to addition of @xmath445 sites to the @xmath63 effective lattices of @xmath63 fermion branches such that @xmath446 . for the @xmath63 effective lattice ,
one local @xmath63 fermion `` occupied site '' refers to @xmath59 sites of the @xmath13-spin ( @xmath246 ) or spin ( @xmath247 ) effective lattice .
alike for the corresponding latter lattice , the @xmath447 sites occupied by deconfined @xmath13-spinons ( @xmath246 ) or deconfined spinons ( @xmath247 ) are among those playing the role of the @xmath63 effective lattice `` unoccupied sites '' .
unlike for the former lattice , it is found below that a number @xmath445 of sites of each @xmath444 fermion with a number @xmath421 of confined @xmath13-spinons ( @xmath246 ) or confined spinons ( @xmath247 ) play as well the role of @xmath63 effective lattice `` unoccupied sites '' .
the conjugate variables of the @xmath63 effective lattice real - space coordinates are the discrete momentum values of the @xmath63 band .
as shown in appendix a , for the 1d model such discrete momentum values are the quantum numbers of the exact ba solution .
for the hubbard model on the square lattice the @xmath64 band momentum discrete values of state representations belonging to the one- and two - electron subspace introduced in section v are good quantum numbers as well . as reported in ref .
@xcite for the @xmath64 fermion operators , the @xmath63 fermion operators can be generated from the operators of corresponding hard - core @xmath63 bond - particle operators as follows , @xmath448 here @xmath449 is the jordan - wigner phase @xcite operator , the indices @xmath450 and @xmath451 refer to sites of the @xmath63 effective lattice , and @xmath452 are the corresponding momentum - dependent @xmath63 fermion operators .
the number @xmath453 of discrete momentum values of the @xmath63 momentum band equals that of sites of the @xmath63 effective lattice .
its expression is derived below in section iv - d . alike in 1d and as illustrated in section vi and appendix d for the @xmath64
bond - particle operators of the model on the square lattice , the @xmath13-spin - neutral @xmath59-@xmath13-spinon composite @xmath60 bond - particle operators and spin - neutral @xmath59-spinon composite @xmath61 bond - particle operators denoted in eq .
( [ f - an - operators ] ) by @xmath454 where @xmath88 are constructed to inherently upon acting onto their @xmath63 effective lattice anticommuting on the same site and commuting on different sites .
hence they are hard - core like and can be transformed onto fermionic operators , as given in that equation .
for @xmath455 the algebra behind their construction in terms of the elementary @xmath13-spinon or spinon operators of eqs .
( [ sir - pir])-([rotated - quasi - spin ] ) is much more cumbersome than that of the two - spinon @xmath64 bond particles studied in section vi and appendix d. fortunately , the only property needed for the goals of this paper is that upon acting onto their @xmath63 effective lattice they are hard - core like .
the expressions given in eq .
( [ f - an - operators ] ) apply to @xmath456 branches provided that @xmath457 .
one can also handle the problem when @xmath241 and @xmath458 for a given @xmath456 branch . then provided that @xmath459 for all remaining @xmath444 branches with a number of @xmath13-spinon or spinon pairs @xmath421 one finds below that @xmath460 .
the momentum of the operators @xmath452 is in that limiting case given by @xmath461 .
for the states that span the corresponding @xmath241 and @xmath458 subspace all sites of the @xmath63 effective lattice are occupied and the @xmath63 momentum band is full .
if @xmath462 is finite one has @xmath463 discrete momentum values @xmath464 compactly distributed around zero momentum .
their cartesian components momentum spacing is @xmath465 .
a case of interest is when @xmath241 and @xmath466 .
then the @xmath63 effective lattice has a single site and the corresponding @xmath63 band a single discrete momentum value , @xmath467 . in that case @xmath468 .
hence @xmath469 and @xmath470 where @xmath467 .
the operators @xmath452 act onto subspaces with fixed values for the set of numbers @xmath85 , @xmath462 , and @xmath471 for @xmath421 branches .
( below it is shown that this is equivalent to fixed values for the set of numbers @xmath28 and @xmath471 for all @xmath472 including @xmath473 . )
such subspaces are spanned by mutually neutral states .
those are states with fixed values for the numbers of @xmath63 fermions and @xmath63 fermion holes .
hence such states can be transformed into each other by @xmath63 band particle - hole processes .
creation of one @xmath63 fermion is a process that involves the transition between two states belonging to different such subspaces .
it is a well - defined process whose generator is the product of two operators .
the first operator may add sites to or remove sites from the @xmath63 effective lattice .
alternatively , it may introduce corresponding changes in the @xmath63 momentum band .
the second operator is the creation operator @xmath474 or @xmath475 appropriate to the excited - state subspace .
provided that @xmath195 is finite for @xmath64 fermions@xcite and @xmath457 as @xmath196 for @xmath456 fermions , the phases @xmath449 given in eq .
( [ f - an - operators ] ) are associated with an effective vector potential @xcite , @xmath476 for the model on the square lattice the vector @xmath477 appearing here is the unit vector perpendicular to the plane .
( often we use units such that the fictitious magnetic flux quantum is given by @xmath478 . ) the components of the microscopic momenta of the @xmath63 fermions are eigenvalues of the two ( and one for 1d ) @xmath63 translation generators @xmath479 and @xmath480 in the presence of the fictitious magnetic field @xmath481 .
that seems to imply that for the model on the square lattice the components @xmath482 and @xmath483 of the microscopic momenta @xmath484 $ ] refer to operators that do not commute .
however , in the subspaces where the operators @xmath452 act onto such are commuting operators .
indeed , those subspaces are spanned by neutral states @xcite . since @xmath485=0 $ ] in such subspaces , for the model on the square lattice
the @xmath63 fermions carry a microscopic momentum @xmath484 $ ] where the components @xmath482 and @xmath483 are well - defined simultaneously . the momentum operator @xmath147 of eq .
( [ p - invariant ] ) commutes with the @xmath63 generators @xmath486 whose eigenvalues are the @xmath63 fermion microscopic momenta @xmath487 .
consistently , within our description it can be written as , @xmath488 in this expression the @xmath57 @xmath13-spinons momentum @xmath151 results from the momentum operator @xmath147 not commuting with the @xmath13-spin off - diagonal generators , the operators @xmath489 and @xmath490 count the numbers @xmath491 $ ] and @xmath492 $ ] of @xmath207 @xmath13-spinons and @xmath13-spinons , respectively , and the @xmath52 and @xmath63 translation generators read , @xmath493\,\hat{n}_{\eta\nu } ( { \vec{q } } ) \ , .
\label{m - generators}\ ] ] here @xmath494 and @xmath495 are the momentum distribution - function operators , @xmath496 respectively .
the @xmath13-spin flip process momentum appearing in the expressions given in the above equations being @xmath151 is consistent with the @xmath13-spin - algebra off - diagonal generators @xmath497 and @xmath498 of eq .
( [ scs ] ) referring to that momentum . that the @xmath60 generators @xmath499 of eq .
( [ m - generators ] ) involve @xmath500 $ ] instead of @xmath501 is consistent as well with the anti - binding character of the @xmath60 fermions reported below .
for the 1d model the form of the momentum operator provided in eq .
( [ p - c - alphanu ] ) is that consistent with the exact solution , as discussed in appendix a. for both the hubbard model on the square and 1d lattice the hamiltonian @xmath138 of eq .
( [ h ] ) and the momentum operator @xmath147 of eqs .
( [ p - invariant ] ) and ( [ p - c - alphanu ] ) obey within the @xmath196 limit the commutation relations , @xmath502 = [ \hat{h},{\hat{m}}_{\alpha\ , , \pm 1/2 } ] = [ \hat{h},{\hat{{\vec{q}}}}_c ] = 0 \ , ; \hspace{0.35 cm } \alpha = \eta , s \
\label{h - commutators}\ ] ] and @xmath503 = [ \hat{{\vec{p}}},{\hat{{\vec{q}}}}_c ] = [ \hat{{\vec{p}}},{\hat{{\vec{q}}}}_{\alpha\nu } ] = 0 \ , ; \hspace{0.35 cm } \alpha = \eta , s \
, , \hspace{0.25 cm } \nu = 1, ... ,\infty \ , .
\label{p - commutators}\ ] ] in turn , the set of commutators @xmath504 $ ] vanish for the 1d model whereas for the model on the square lattice one has in general that @xmath504\neq 0 $ ] .
it follows from the momentum operator expression ( [ p - c - alphanu ] ) that the corresponding momentum eigenvalues @xmath505 can be expressed as a sum of the filled @xmath52 and @xmath63 fermion microscopic momenta . for most subspaces the maximum value @xmath506 of the @xmath63 fermion @xmath13-spinon ( @xmath214 ) or spinon ( @xmath215 ) pair number @xmath473 behaves as @xmath507 in the @xmath196 limit . in that case
the corresponding set of @xmath63 translation generators @xmath486 of eq .
( [ m - generators ] ) is infinite , @xmath508 .
we emphasize that as given in eq .
( [ h - commutators ] ) , the commutator @xmath509 $ ] vanishes .
indeed , for both the hubbard model on the square and 1d lattice the @xmath52 band momenta are good quantum numbers for the whole hilbert space .
as further discussed below , the conservation of such momenta is related to both the local separation of the electronic degrees of freedom due to the @xmath48 local @xmath19 gauge symmetry and the unitarity of the operator @xmath89 . for the square - lattice quantum
liquid , whose construction performed in section v and ref .
@xcite relies on the general description introduced in this paper , both the @xmath52 band and @xmath64 band discrete momentum values are good quantum numbers .
the shape of the @xmath64 momentum band and of its boundary as well as the form of the @xmath64 and @xmath52 fermion energy dispersions are problems addressed in ref .
@xcite for the hubbard model on the square lattice in the one- and two - electron subspace .
each subspace spanned by states with fixed values of @xmath28 , @xmath26 , and @xmath27 and hence also with fixed values of @xmath238 and @xmath510 can be divided into smaller subspaces spanned by states with fixed values for the set of numbers @xmath511 and @xmath512 , which must obey the sum rules of eq .
( [ m - l - sum ] ) .
the numbers @xmath511 and @xmath512 correspond to operators @xmath513 and @xmath514 that commute with both the momentum operator of eqs . ( [ p - invariant ] ) and ( [ p - c - alphanu ] ) and the seven generators given in eqs .
( [ or - ope ] ) and ( [ scs ] ) of the group @xmath154/z_2=so(3)\times so(3)\times u(1)$ ] associated with the model global symmetry . in turn
, the lack of an infinite set of conservation laws implies that for the model on the square lattice the operators @xmath513 and @xmath514 do not commute in general with the hamiltonian . hence for that model the numbers @xmath511 and @xmath512 are not in general good quantum numbers .
the same applies to the set of @xmath63 translation generators @xmath486 of eq .
( [ m - generators ] ) in the presence of the fictitious magnetic field @xmath515 of eq .
( [ a - j - s1 - 3d ] ) .
as given in eqs .
( [ h - commutators ] ) and ( [ p - commutators ] ) , they commute with the momentum operator yet in general do not commute with the hamiltonian of the model on the square lattice . for 1d both such @xmath63 translation generators and the number operators of the set @xmath516 commute with the hamiltonian . within the @xmath196 limit
there is an infinite number of such generators and numbers . as discussed in appendix a , they are associated with the set of infinite conservation laws behind the model integrability @xcite . for the hubbard model on the square lattice in the one- and two - electron subspace
most of such numbers vanish and a few of them become good quantum numbers .
that the momentum operator commutes with the electron - rotated - electron unitary operator @xmath89 implies that the momentum eigenvalues of the states @xmath105 studied below in section iv - e are independent of @xmath10 .
they are the same as those of the corresponding states @xmath517 .
for both the hubbard model on the square and 1d lattice it follows from eqs .
( [ p - c - alphanu ] ) and ( [ m - generators ] ) that such momentum eigenvalues are given by , @xmath518\,n_{\eta\nu } ( { \vec{q } } ) + \vec{\pi}\,m_{\eta\ , , -1/2 } \ , .
\label{p-1 - 2-el - ss}\ ] ] the momentum distribution functions @xmath519 and @xmath520 appearing here are such that application of the corresponding momentum distribution - function operators ( [ nc - s1op ] ) onto the momentum eigenstates @xmath107 gives @xmath521 and @xmath522 , respectively .
indeed @xmath519 and @xmath520 are eigenvalues of such operators that read @xmath523 and @xmath524 for filled and unfilled momentum values , respectively . for the hubbard model on the 1d lattice the validity of expression ( [ p-1 - 2-el - ss ] ) is confirmed by the exact solution , the @xmath52 and @xmath63 fermion discrete momentum values being good quantum numbers .
hence as discussed in appendix a , for 1d the momentum distribution - function operators @xmath494 and @xmath495 of eq .
( [ nc - s1op ] ) commute with the hamiltonian .
that the distributions @xmath525 and @xmath526 are given by @xmath523 or @xmath524 does not imply that the same applies to the rotated - electron momentum distribution @xmath527 where @xmath528 .
for @xmath8 such a rotated - electron momentum distribution is independent of @xmath10 , equaling the @xmath529 electron momentum distribution .
however , only for the @xmath146 and @xmath12 ground state of both the model on the 1d and square lattice does the rotated - electron momentum distribution simply reads @xmath530 , alike @xmath519 . for the model on the 1d lattice the corresponding @xmath146 , @xmath12 , and @xmath529 electron momentum distribution @xmath531 is plotted in fig .
3 ( a ) of ref . @xcite . in turn , for @xmath145 and @xmath12 ground states the rotated - electron momentum distribution is not merely given by @xmath523 or @xmath524 , being a function of the rotated - electron momentum @xmath532 .
it has contributions both from the @xmath52 fermion and spinon degrees of freedom . for the 1d model
it is studied in ref .
@xcite in the @xmath529 limit for electrons , on combining the exact ba ground - state wave function with numerical methods .
( the distribution of that reference applies as well to rotated electrons for @xmath8 . ) that the energy spectrum depends on @xmath10 is for the hubbard model on the square lattice behind the @xmath145 and @xmath12 ground states belonging in general to different @xmath103 towers for different values of @xmath10 .
the exception is the @xmath146 and @xmath12 ground state , which belongs to the same @xmath103 tower for all @xmath8 values @xcite . in the following
we provide further information on why for the model on the square lattice the @xmath52 band momenta are conserved and the momentum operator and corresponding eigenvalues have the form given in eqs .
( [ p - c - alphanu ] ) and ( [ p-1 - 2-el - ss ] ) , respectively . in the @xmath529 limit
electron single and doubly occupancy become good quantum numbers and the electronic degrees of freedom contributing to the momentum decouple into two main types : i ) those associated with electron hopping , which contribute to the kinetic energy ; ii ) those associated with the @xmath59-@xmath13-spinon composite @xmath60 fermions and @xmath59-spinon composite @xmath61 fermion occupancy configurations , which do not contribute to the kinetic energy .
the discrete momentum values associated with electron hopping and the finite kinetic energy are carried by quasicharge particles whose operator @xmath533 is given by , @xmath534 here within the notation of ref .
@xcite , @xmath535 are the real - space coordinates of the quasicharge particles of that reference .
such particles are the @xmath52 fermion holes in that limit .
the momentum values ( ii ) refer to the composite @xmath60 and @xmath61 fermions whose spectrum is dispersionless for @xmath529 . as discussed below in section iv - c ,
such objects are anti - binding and binding neutral superpositions of @xmath435 pseudospins ( or @xmath13-spinons ) and @xmath435 spins ( our spinons ) , respectively .
the corresponding local spin and pseudospin operators have expressions in terms of electron operators similar to those of eqs .
( [ sir - pir])-([rotated - quasi - spin ] ) of the corresponding spinon and @xmath13-spinon operators , respectively , in terms of rotated - electron operators @xcite . in the @xmath48
limit the @xmath52 fermion momentum is the part of the corresponding electronic momentum associated with the kinetic energy .
indeed the electronic momentum has in addition contributions from the spin occupancy configurations . in the @xmath48 limit
such configurations are degenerate and have vanishing energy . note that in the @xmath48 limit the electronic momentum is not a good quantum number .
however , the part of it associated with the kinetic energy alone , which results from the electronic motion of the singly occupied sites relative to the remaining sites , is .
the above separation of the electronic degrees of freedom contributing to the momentum follows from in the @xmath48 limit the hubbard model on a bipartite lattice having a local @xmath19 gauge symmetry , stronger than the related global @xmath20/z_2 ^ 2=[so(4)\times u(1)]/z_2=so(3)\times so(3)\times u(1)$ ] symmetry @xcite .
such a local gauge symmetry has indeed stronger consequences in terms of the electronic degrees of freedom separation than a global symmetry alone .
( in that limit the finite - electron - double occupancy @xmath13-spin configurations are degenerate but refer to infinite energy and thus are not accessible to the finite energy physics . )
importantly , the finite - energy electronic degrees of freedom associated with the momentum values contributing to the kinetic energy and not contributing to it are associated with the @xmath52 fermion local @xmath5 gauge symmetry and local spin @xmath11 gauge symmetry , respectively , of the model local @xmath19 gauge symmetry .
such a local separation of the electronic degrees of freedom associated with the momentum contributions arising from the kinetic energy and spin configurations , respectively , which follows from two independent local gauge symmetries , is behind the @xmath52 fermion momentum values being good quantum numbers of the model on the square lattice in the @xmath48 limit .
the point is that while the electronic momentum values have contributions both from the @xmath52 fermion and spin degrees of freedom so that the electrons are not true non - interacting spinless fermions , in the @xmath48 limit the @xmath52 fermions are .
the unitarity of the operator @xmath89 then assures that such momentum values are good quantum numbers of the model on the square lattice for @xmath8 as well .
this is why , as given in eqs .
( [ h - commutators ] ) and ( [ p - commutators ] ) , for the whole interaction range @xmath8 both the hamiltoninan and momentum operator commute with the @xmath52 translation generator @xmath536 of eq . ( [ m - generators ] ) . and
this applies both to the model on the 1d and square lattice .
for @xmath8 the distribution @xmath537 is then an eigenvalue of the operator @xmath538 of eq .
( [ nc - s1op ] ) .
it reads @xmath523 and @xmath524 for filled and unfilled momentum values , respectively .
since for @xmath8 the set of momentum eigenstates @xmath104 introduced below in section iv - e is complete , the corresponding energy eigenstates @xmath272 of the hubbard model on the square lattice can be expressed as superpositions of sub - sets of the former states with the same values for the distribution @xmath537 . concerning the @xmath64 fermion momentum values , it turns out that the @xmath48 local spin @xmath11 gauge symmetry also implies that they are good quantum numbers of the hubbard model on the square lattice provided that both the corresponding numbers @xmath539 and @xmath540 of unfilled and filled discrete momentum values , respectively , are conserved .
this condition is not in general fulfilled for that model .
fortunately , it is fulfilled for it in the one- and two - electron subspace as defined below in section v. the unitarity of the operator @xmath89 then assures that in such a subspace the @xmath64 band momentum values are conserved for @xmath8 as well .
the @xmath8 model global @xmath3 symmetry can be written as @xmath154/z_2 $ ] . both for the hubbard model on the square and 1d lattices
it is useful to rewrite the momentum eigenvalues @xmath505 of eq .
( [ p-1 - 2-el - ss ] ) in terms of two contributions @xmath541 and @xmath542 arising from the hidden @xmath5 symmetry and @xmath17 symmetry , respectively , state representation contributions . for both such models the momentum contribution @xmath543 to the momentum eigenvalues
, @xmath544 is for @xmath8 a good quantum number .
this simply follows from @xmath505 and @xmath541 being good quantum numbers as well . in expression ( [ p-1 - 2-el - ss-2d ] ) @xmath541 , @xmath545 , and @xmath546 are the momentum contributions arising from the @xmath52 fermion , @xmath13-spinon , and spinon occupancy configurations , respectively .
the latter two contributions read @xmath547\,n_{\eta\nu } ( { \vec{q}})+\vec{\pi}\,m_{\eta\ , , -1/2}$ ] and @xmath548 , respectively . while the @xmath17 momentum @xmath549 given in eq .
( [ p-1 - 2-el - ss-2d ] ) is a good quantum number , the operators associated with @xmath545 and @xmath546 and the corresponding sets of translation generators @xmath499 and @xmath550 , respectively , of eq . ( [ m - generators ] ) labelled by the index @xmath551
do not commute in general with the hamiltonian of the hubbard model on the square lattice , unlike for 1d . for @xmath48
all @xmath13-spin and spin configurations are degenerate and do not contribute to the kinetic energy .
nevertheless they lead to overall momentum contributions @xmath545 and @xmath546 , respectively . as mentioned above , an important point is that the @xmath63 fermion operators are defined in and act onto subspaces spanned by mutually neutral states . for the hubbard model on the square lattice in these subspaces
the @xmath63 translation generators @xmath552 and @xmath553 commute .
this implies that the @xmath63 translation generators @xmath486 of eq .
( [ m - generators ] ) in the presence of the fictitious magnetic field @xmath515 of eq .
( [ a - j - s1 - 3d ] ) commute with the momentum operator both in the @xmath48 limit and for @xmath10 finite , as given in eq . ( [ p - commutators ] ) .
the momentum operator can then be written as in eqs .
( [ p - c - alphanu ] ) and ( [ m - generators ] ) both for the hubbard model on the 1d and square lattices .
the exact expressions of the numbers of @xmath63 band discrete momentum values @xmath453 are found below in section iv - d .
thus for the model on the square lattice the momentum area @xmath554 ^ 2 $ ] and discrete momentum values number @xmath555 of the @xmath63 bands are known .
in contrast , the shape of their boundary remains in general an open issue .
the shape of the @xmath52 band is the same as that of the first brillouin zone .
however , the discrete momentum values may have small overall shifts under transitions between subspaces with different @xmath556 values . the studies of ref .
@xcite use several approximations to obtain useful information on the @xmath52 and @xmath64 bands momentum values . in that reference the corresponding @xmath52 and @xmath64 fermion energy dispersions and velocities
are studied for the hubbard model on the square lattice in the one- and two - electron subspace , for which both the translation generators @xmath536 and @xmath557 of eq .
( [ m - generators ] ) commute with that model hamiltonian . since the microscopic momenta of the @xmath52 fermions are good quantum numbers both for the hubbard model on the square and 1d lattices , one may define an energy dispersion @xmath558 @xcite . for the 1d model also
the @xmath63 energy dispersions @xmath559 are well defined . in turn , for the model on the square lattice an energy dispersion @xmath560 is for @xmath456 branches well defined for the momentum values for which such objects are invariant under the electron - rotated - electron unitary transformation . for general @xmath63 fermion momentum values
there is not in general a dispersion @xmath561 defining a one - to - one correspondence between the energy @xmath562 and momentum @xmath487 .
however , the range of the @xmath63 fermion energy @xmath562 remains well defined .
a @xmath64 fermion energy dispersion @xmath563 is well defined for the square - lattice quantum liquid @xcite .
the quantum - object occupancy configurations of the ground state are found below .
they are consistent with the energies for creation onto such states of our description quantum objects .
for instance , the elementary energies @xmath564 and @xmath565 of eq .
( [ energies ] ) of appendix b correspond to creation onto a @xmath566 and @xmath567 ground state of a deconfined @xmath57 spinon and a deconfined @xmath57 @xmath13-spinon , respectively . here
@xmath568 is the bohr magneton , @xmath569 the magnetic field , and @xmath332 the chemical potential . the energy @xmath564 ( and @xmath565 ) refers to an elementary spin - flip ( and @xmath13-spin - flip ) process .
it transforms a deconfined @xmath58 spinon ( and deconfined @xmath58 @xmath13-spinon ) into a deconfined @xmath57 spinon ( and deconfined @xmath57 @xmath13-spinon ) .
such elementary energies control the range of several physically important energy scales . within our lws representation ,
a deconfined @xmath58 spinon ( and deconfined @xmath58 @xmath13-spinon ) has vanishing energy so that @xmath570 ( and @xmath571 ) .
it follows that the energy of a pair of deconfined spinons ( and deconfined @xmath13-spinons ) with opposite projections is @xmath572 ( and @xmath402 ) . indeed due to the invariance of such objects under the electron - rotated - electron unitary transformation , they are not energy entangled and the total energy is the sum of their individual energies . in the following we confirm that ground states have no @xmath60 fermions and no @xmath61 fermions with @xmath573 spinon pairs . the corresponding energies @xmath574 and @xmath575 , respectively , considered below refer to creation onto the ground state of one of such objects .
we start by providing a set of useful properties .
we emphasize that some of these properties are not valid for descriptions generated by rotated - electron operators associated with the general unitary operators @xmath121 considered in ref .
the following properties rather refer to the specific operator description associated with the rotated - electron operators @xmath576 of eq .
( [ rotated - operators ] ) as defined in section ii .
some of the following results are obtained from extension to the model on the square lattice of exact results extracted from the 1d model ba solution .
however such an extension accounts for the different physics of such models . both for the model on the square and 1d lattice ,
the range of the energy @xmath562 for addition onto the ground state of one @xmath63 fermion derived below is that consistent with the interplay of the transformation laws of the @xmath63 fermions under the electron - rotated - electron unitary transformation with the model global @xmath3 symmetry . alike for 1d , for the model on the square lattice one @xmath60 fermion is a @xmath13-spin - neutral anti - binding configuration of a number @xmath62 of confined @xmath57 @xmath13-spinons and an equal number of confined @xmath58 @xmath13-spinons .
symmetry implies that for @xmath8 there is no energy overlap between the @xmath574 ranges corresponding to different @xmath62 branches .
fermions belonging to neighboring @xmath60 and @xmath577 branches differ in the number of @xmath13-spinon pairs by one .
the requirement for the above lack of energy overlap is then that the energy bandwidth of the @xmath574 range is smaller than or equal to @xmath329 . for all @xmath194 values , the energy scale @xmath578 $ ] where @xmath579 at @xmath146 refers to the energy of a pair of deconfined @xmath13-spinons of opposite @xmath13-spin projection .
such properties imply the following range for the energy @xmath574 , @xmath580 where @xmath579 at @xmath146 .
deconfined @xmath13-spinons are invariant under the electron - rotated - electron unitary transformation .
thus they are non interacting and their energies are additive .
consistently , for all @xmath194 values @xmath581 is the energy of @xmath473 deconfined @xmath57 @xmath13-spinons and @xmath473 deconfined @xmath58 @xmath13-spinons .
for instance , for @xmath145 the energy @xmath582 is as well that for creation of a number @xmath62 of deconfined @xmath57 @xmath13-spinons onto a @xmath583 ground state with @xmath59 deconfined @xmath58 @xmath13-spinons .
such a creation refers to @xmath473 @xmath13-spin - flip processes ( transformation of @xmath473 rotated - electron unoccupied sites into @xmath473 rotated - electron doubly occupied sites . ) for @xmath12 and @xmath145 the number @xmath584 decreases continuously for increasing values of @xmath10 .
it has the limiting behaviors @xmath585 for @xmath586 and @xmath587 for @xmath48 .
hence the @xmath574 range vanishes for @xmath48 .
the latter behavior is associated with the full degeneracy of the @xmath13-spin configurations reached as @xmath48 . in turn , at @xmath12 and @xmath146 the number @xmath584 vanishes and @xmath588 for the whole finite interaction range @xmath8 . as discussed below
, this behavior follows from the invariance under the electron - rotated - electron unitary transformation of a @xmath60 fermion created onto a @xmath146 and @xmath12 ground state .
a @xmath61 fermion is a spin - neutral binding configuration of a number @xmath62 of confined @xmath57 spinons and an equal number of confined @xmath58 spinons .
again , symmetry implies that for @xmath8 there is no energy overlap between the @xmath575 ranges of different @xmath62 branches . for @xmath61 branches with a number of spinon pairs
@xmath573 such an energy range bandwidth is for the present binding configurations and for the same reasoning as for the @xmath60 fermion smaller than or equal to @xmath589 .
for all @xmath43 values , @xmath590 $ ] equals the energy of a pair of deconfined spinons of opposite spin projection .
hence the range of the energy @xmath575 for addition onto the ground state of one @xmath61 fermion with @xmath573 spinon pairs is , @xmath591 deconfined spinons are invariant under the electron - rotated - electron unitary transformation .
thus their energies are additive .
it follows that for all @xmath43 values @xmath592 is the energy of @xmath473 deconfined @xmath57 spinons and @xmath473 deconfined @xmath58 spinons .
for example , for @xmath320 the energy @xmath593 is that for creation of a number @xmath62 of deconfined @xmath57 spinons onto a @xmath594 ground state with @xmath59 deconfined @xmath58 spinons .
such a creation refers to @xmath473 spin - flip processes .
the number @xmath595 decreases continuously for increasing values of @xmath10 . for any fixed @xmath43 value it has the limiting behavior @xmath596 for @xmath586 .
moreover , it is such that @xmath597 and @xmath598 for @xmath48 .
hence the energy bandwidth of the @xmath575 range vanishes for @xmath48 .
such a behavior is associated with the full degeneracy of the spin configurations reached for @xmath48 .
note that at @xmath12 and thus @xmath599 one has that @xmath600 for the whole finite interaction range @xmath8 .
this behavior follows from the invariance under the electron - rotated - electron unitary transformation of a @xmath61 fermion with @xmath455 spinon pairs created onto a @xmath12 ground state . in turn
, it is found below that for a @xmath12 and @xmath601 ground state all sites of the @xmath64 effective lattice are occupied .
hence the corresponding @xmath64 momentum band is full .
the range of the energy @xmath602 for removal from that state of one @xmath64 fermion then is , @xmath603 here @xmath604 denotes the @xmath64 fermion pairing energy per spinon considered in ref . @xcite . for small finite hole concentrations
@xmath605 it vanishes both in the @xmath586 and @xmath606 limits and goes through a maximum value at @xmath607 . at fixed @xmath10 values
it decreases for increasing @xmath194 and vanishes for @xmath608 . here
@xmath609 for @xmath610 is a critical hole concentration below which the @xmath64 fermion pairing refers to a spin short - range order @xcite .
for @xmath8 and @xmath146 its magnitude @xmath611 is larger than for @xmath379 and the @xmath64 fermion spinon pairing refers to an antiferromagnetic long - range order @xcite . on the other hand , at @xmath12
the energy scale @xmath612 is the @xmath64 fermion energy nodal bandwidth defined in ref .
its maximum magnitude is reached at @xmath613 .
for @xmath8 it decreases monotonously for increasing values of @xmath10 , vanishing for @xmath48 .
that @xmath614 for @xmath48 is associated with the full degeneracy of the spin configurations reached in that limit . in it the spectrum of the two - spinon composite @xmath64 fermions becomes dispersionless for the square - lattice quantum liquid of ref .
@xcite . in the @xmath48
limit the @xmath64 fermion occupancy configurations that generate the spin degrees of freedom of spin - density @xmath12 ground states considered below become for the 1d model those of the spins of the spin - charge factorized wave function introduced by woynarovich @xcite .
it is associated with the @xmath271-electron amplitude @xmath286 of eq .
( [ amplitude-1d - rot - elec ] ) , which was later rediscovered by ogata and shiba @xcite . in turn , for the model on the square lattice such configurations become in that limit and within a mean - field approximation for the fictitious magnetic field @xmath615 of eq .
( [ a - j - s1 - 3d ] ) for @xmath616 , those of a full lowest landau level with @xmath617 one-@xmath64-fermion degenerate states of the @xmath618 quantum hall effect @xcite . here
@xmath619 is the number of both sites of the @xmath64 effective square lattice and @xmath64 band discrete momentum values . for finite @xmath10 values and @xmath145 the degeneracy of the @xmath620 one-@xmath64 fermion states of the square - lattice quantum liquid studied in ref .
@xcite is removed by the emergence of a finite - energy - bandwidth @xmath64 fermion dispersion .
however , the number of @xmath64 band discrete momentum values remains being given by @xmath621 .
in addition , the @xmath64 effective lattice spacing remains reading @xmath622 . here
@xmath623 is the fictitious - magnetic - field length .
the energy @xmath624 for addition onto the ground state of one @xmath52 fermion of a given momentum and the energy @xmath625 for removal from that state of such a @xmath52 fermion have the following ranges , @xmath626 \ , ; \hspace{0.5 cm } 0 \leq -\epsilon_{c}\leq w_c^p \ , , \hspace{0.25 cm } d = 1,2 \ , , \label{energy - c}\ ] ] respectively .
here @xmath627\in ( 0,4dt)$ ] increases monotonously for increasing values of hole concentration @xmath330 .
the energy bandwidth @xmath628 depends little on @xmath10 .
for @xmath8 it has the following limiting behaviors , @xmath629 the behaviors reported here for @xmath624 are justified in ref .
the minimum magnitude of the energy @xmath338 for creation of a number @xmath630 of rotated - electron doubly occupied sites onto a @xmath12 and @xmath601 ground state given in eq .
( [ min - d ] ) may be expressed in terms of both the numbers of deconfined @xmath13-spinons and @xmath60 fermions as follows , @xmath631 here @xmath632 is the energy scale whose limiting behaviors are given in eq .
( [ dmh ] ) .
consistently , the following @xmath13-spinon and @xmath60 fermion energy magnitudes hold , @xmath633 the corresponding energy magnitudes concerning creation of deconfined spinons and @xmath61 fermions with @xmath455 spinon pairs onto @xmath601 ground states with arbitrary values of @xmath43 read , @xmath634 hence for all hole concentrations @xmath194 and all spin densities @xmath43 the inequalities @xmath635 = 2\nu\vert\mu\vert$ ] where @xmath636 at @xmath146 and @xmath637 = 2\nu\mu_b\,\vert h\vert$ ] , respectively , hold . furthermore , @xmath638= 2\vert\mu\vert$ ] where again @xmath579 at @xmath146 and @xmath639= 2\mu_b\,\vert h\vert$ ] .
a related important property is that @xmath60 fermions of any @xmath62 branch and @xmath61 fermions with @xmath573 spinon pairs whose energy obeys the following relations , @xmath640 = 2\nu\vert\mu\vert \ , , \hspace{0.25 cm } \nu = 1,2, ... \hspace{0.2cm}{\rm for}\hspace{0.2cm}x\neq 0 \nonumber \\ & = & \nu [ \epsilon_{\eta,-1/2 } + \epsilon_{\eta,+1/2 } ] = 2\nu\mu^0 \ , , \hspace{0.25 cm } \nu = 1,2, ...
\hspace{0.2cm}{\rm at}\hspace{0.2cm}x=0 \nonumber \\ \epsilon_{s\nu } & = & \nu [ \epsilon_{s,-1/2 } + \epsilon_{s,+1/2 } ] = 2\nu\mu_b\,\vert h\vert \ , , \hspace{0.25 cm } \nu = 2,3 , ... \ , , \label{invariant - v}\end{aligned}\ ] ] remain invariant under the electron - rotated - electron unitary transformation .
those are non - interacting objects such that their energy is additive in the individual energies of the corresponding @xmath59 @xmath13-spinons and @xmath59 spinons , respectively .
therefore , for @xmath8 they refer to the same occupancy configurations in terms of both rotated electrons and electrons .
this refers to the @xmath13-spin and spin degrees of freedom , respectively , of such rotated - electron occupancy configurations .
the corresponding rotated - electron degrees of freedom associated with the @xmath52 fermion @xmath5 symmetry are not in general invariant under that transformation .
another important property is that a @xmath61 fermion with @xmath455 spinon pairs ( and a @xmath60 fermion ) created onto an initial @xmath641 and @xmath599 ( and @xmath642 ) ground state remains invariant under the electron - rotated - electron unitary transformation .
indeed and as reported above , such an object has a uniquely defined energy @xmath600 ( and @xmath588 ) .
furthermore , its momentum is @xmath643 ( and @xmath644 ) .
hence it obeys indeed to the condition ( [ invariant - v ] ) . in turn ,
@xmath61 fermions with @xmath455 spinon pairs ( and @xmath60 fermions ) created onto initial @xmath645 ( and @xmath646 ) ground states may have energy @xmath575 smaller than @xmath592 ( and energy @xmath574 larger than @xmath581 ) , so that they are not necessarily invariant under the electron - rotated - electron unitary transformation . to understand this property , we consider a set of @xmath59 deconfined @xmath13-spinons ( and @xmath59 deconfined spinons ) including @xmath473 deconfined @xmath58 @xmath13-spinons ( and @xmath473 deconfined @xmath58 spinons ) and @xmath473 deconfined @xmath57 @xmath13-spinons ( and @xmath473 deconfined @xmath57 spinons ) .
such a set of @xmath59 deconfined @xmath13-spinons ( and @xmath59 deconfined spinons ) refers to a @xmath647 and @xmath648 @xmath13-spin multiplet configuration of energy @xmath581 where @xmath636 at @xmath146 ( and a @xmath649 and @xmath650 spin multiplet configuration of energy @xmath592 ) . in turn , the @xmath59 @xmath13-spinons ( and spinons ) of a @xmath60 fermion ( and @xmath61 fermion ) correspond to a @xmath651 @xmath13-spin singlet configuration of energy @xmath652 ( and a @xmath653 spin singlet configuration of energy @xmath654 ) .
the point is that only when the energy @xmath574 ( and @xmath575 ) of a @xmath60 fermion ( and @xmath61 fermion ) @xmath59-@xmath13-spinon @xmath13-spin singlet configuration ( and @xmath59-spinon spin singlet configuration ) reads @xmath655 ( and @xmath656 ) , as given in eq .
( [ invariant - v ] ) , is it degenerated with that of the above @xmath59-deconfined @xmath13-spinon @xmath13-spin multiplet configuration ( and @xmath59-deconfined spinon spin multiplet configuration ) . only when that occurs is such a @xmath60 fermion ( and @xmath61 fermion ) invariant under the electron - rotated - electron unitary transformation .
on the other hand , @xmath60 fermions of any @xmath62 branch and @xmath61 fermions with @xmath573 spinon pairs whose energies obey the inequalities @xmath657 and @xmath658 , respectively , are not invariant under the electron - rotated - electron transformation . furthermore ,
for finite @xmath10 values @xmath52 fermions and @xmath64 fermions are not in general invariant under that transformation . for the 1d model
the @xmath63 fermion energy @xmath559 depends on the @xmath63 fermion momentum @xmath659 , which as discussed in appendix a is a good quantum number . for @xmath8 and
the @xmath61 branches with @xmath455 spinon pairs and all @xmath60 branches such momenta belong to the range @xmath660 and @xmath661 , respectively . only at the limiting momenta @xmath662 ( and @xmath663 )
is the invariance condition ( [ invariant - v ] ) met by the @xmath60 fermion energy @xmath664 ( and @xmath61 fermion energy @xmath665 for @xmath455 branches ) .
the @xmath60 fermion energy dispersions @xmath664 are plotted for the @xmath424 ( and @xmath666 ) branches in figs .
8 ( a ) and 9 ( a ) ( and figs . 8 ( b ) and 9 ( b ) ) of ref .
@xcite as a function of @xmath659 for several @xmath10 values and electronic density @xmath667 and @xmath668 , respectively .
( in that reference the @xmath60 fermions are called @xmath669 pseudoparticles . ) the zero - energy level of figs . 8 ( a ) and 9 ( a ) ( and figs .
8 ( b ) and 9 ( b ) ) refers to the energy @xmath329 ( and @xmath670 ) of the invariance condition ( [ invariant - v ] ) for @xmath671 ( and @xmath672 ) . for the hubbard model on the square lattice
the @xmath63 fermions whose energy obeys the invariance condition ( [ invariant - v ] ) have a well - defined momentum @xmath673 , which can point in different directions . at @xmath146 ( and @xmath12 )
the momentum @xmath674 ( and @xmath675 ) vanishes .
this is alike for the above corresponding momentum @xmath676 ( and @xmath677 ) of the 1d model , which vanishes at @xmath146 ( and @xmath12 ) . while at 1d the momentum @xmath676 ( and @xmath677 ) can for @xmath145 ( and @xmath320 )
have two values , for the square - lattice model the momentum @xmath674 ( and @xmath675 ) can for @xmath145 ( and @xmath320 ) point to several directions .
for instance , for @xmath678 and @xmath12 where @xmath679 is the hole concentration considered in ref .
@xcite below which the ground - state @xmath64 fermion spinon pairing is for @xmath145 associated with a short - range spin order , a good approximation for the momentum @xmath674 centered at @xmath680 $ ] is , @xmath681 here @xmath682 is the @xmath52 fermion hole fermi momentum considered in ref .
@xcite and @xmath683 is a unit vector centered at @xmath684 of cartesian components @xmath685 $ ] .
as required , @xmath686 as @xmath379 . consistently with the energy values given in eqs .
( [ energy - eta ] ) and ( [ energy - s ] ) , one finds that both for the model on the 1d and square lattice in the subspace spanned by the @xmath145 and @xmath320 lws ground states and their excited energy eigenstates of energy @xmath687 the @xmath13-spinon and spinon numbers are given by @xmath688 , @xmath689 and @xmath690 , @xmath691 , respectively , so that @xmath692 . hence for such energy eigenstates
the numbers of @xmath60 fermions and @xmath420 fermions with @xmath693 spinon pairs vanish and the number of @xmath52 fermions is @xmath694 and that of @xmath64 fermions is conserved and reads @xmath695 . except that the @xmath57 @xmath13-spinons and @xmath57 spinons play the role of the @xmath58 @xmath13-spinons and @xmath58 spinons , respectively , and vice versa , similar results are reached for hwss .
comparison of the occupancies of the spin lws ground states ( @xmath320 ) and spin hws ground states ( @xmath322 ) provides useful information . from it and again consistently with the energy values of eqs .
( [ energy - eta ] ) and ( [ energy - s ] ) , one finds that a @xmath12 ground state for which @xmath271 is even and @xmath601 has @xmath696 and @xmath697 values as given above whereas @xmath698 so that @xmath699 and @xmath700 . a local @xmath60 ( and @xmath61 ) fermion refers to a well - defined superposition of @xmath13-spinon ( and spinon ) occupancy configurations involving @xmath435 sites of the @xmath13-spin ( and spin ) effective lattice .
for the model on the square lattice in the one- and two - electron subspace of most physical interest the only existing @xmath63 effective lattice is the @xmath64 effective lattice .
( in it the @xmath436 and @xmath424 effective lattices either do not exist or have a single occupied site . )
the number of sites , @xmath701 , of the @xmath64 effective lattice plays a major role in the studies of that model one - electron and spin spectra @xcite . alike
all @xmath63 effective lattices , such an effective lattice is exotic , since the general expression for its number of sites depends on the values of @xmath28 , @xmath27 , and set of numbers @xmath512 .
hence the @xmath701 value is subspace dependent . in order to derive the expression of such
a dependence one needs though to solve the general problem referring to all @xmath63 effective lattices .
indeed the values of the set @xmath702 of numbers of sites of such lattices depend on each other .
our goal here is the thus derivation of the expression of the number of @xmath63 effective lattice sites @xmath453 in terms of @xmath28 , @xmath27 , and set of numbers @xmath471 where @xmath472 .
the real - space coordinate @xmath160 of a local @xmath63 fermion plays the role of its center of mass " .
the corresponding index @xmath703 enumerates the sites of the @xmath63 effective lattice whose @xmath453 real - space coordinates have well - defined different values .
fortunately , within the present @xmath33 limit , ( i ) the internal structure of a local @xmath63 fermion and ( ii ) its position specified by the corresponding real - space coordinate @xmath160 are separated problems .
the present analysis refers only to the problem ( ii ) .
concerning the internal structure of a local @xmath63 fermion , the only issue that matters for the present analysis is that the @xmath704 sites of the @xmath13-spin ( and spin ) effective lattice occupied by a given local @xmath60 ( and @xmath61 ) fermion correspond to @xmath435 sites of the original lattice that are not simultaneously occupied by any other composite fermions .
the direct correspondence and relationship of the rotated - electron occupancy configurations to those of the @xmath52 fermions , @xmath63 fermion , deconfined @xmath13-spinons , and deconfined spinons has been constructed to inherently such a property holding .
the number of sites @xmath453 of the @xmath63 effective lattice is an integer number .
however , for the model on the square lattice for which @xmath37 the related number @xmath705 such that @xmath706 is not in general an integer . within the present @xmath367 limit , we consider that it is the integer number closest to it .
this is why in that limit we use the notation @xmath555 for the @xmath63 effective square lattice number of sites . the @xmath63 effective lattice and its @xmath453 sites are well - defined concepts in a subspace in which the values of the set of numbers @xmath51 and @xmath471 with @xmath707 and @xmath472 remain fixed .
( as confirmed below , this is equivalent to the @xmath13-spin @xmath26 ( @xmath246 ) or spin @xmath211 ( @xmath247 ) and values of the set of numbers @xmath471 with @xmath707 and @xmath708 remaining fixed . )
one occupied site of the @xmath63 effective lattice corresponds to @xmath709 sites of the @xmath13-spin ( @xmath214 ) or spin ( @xmath215 ) effective lattice .
hence the @xmath462 local fermions belonging to the same @xmath63 branch occupy @xmath462 and @xmath710 sites of the @xmath63 effective lattice and @xmath13-spin ( @xmath214 ) or spin ( @xmath215 ) effective lattice , respectively . which sites of the @xmath13-spin ( and spin ) effective lattice play the role of unoccupied sites of the @xmath60 ( and @xmath61 ) effective lattice is an issue fully determined by the number of @xmath13-spin - singlet ( and spin - singlet ) configurations of each subspace with fixed values of @xmath28 , @xmath26 , and @xmath27
. those are subspaces of the larger subspace that the @xmath711 vacuum of eq .
( [ vacuum ] ) refers to .
the dimension of such subspaces equals the product of the corresponding numbers of state representations of the @xmath52 fermion hidden @xmath5 symmetry , @xmath13-spin @xmath712 symmetry ( @xmath214 ) , and spin @xmath712 symmetry ( @xmath215 ) , respectively .
those are the three symmetries contained in the model global @xmath6/z_2 ^ 2 $ ] symmetry .
the number of sites of the @xmath63 effective lattice , which equals that of @xmath63 band discrete momentum values , has the general form , @xmath713 \ , , \label{n*}\ ] ] where @xmath714 is the number of unoccupied sites whose expression we derive in the following . to achieve such a goal , we follow the procedures of ref .
@xcite and divide the hilbert space of the model ( [ h ] ) in a set of subspaces spanned by the states with fixed values of @xmath28 , @xmath26 , and @xmath27 and hence also of @xmath51 , @xmath715 $ ] , and @xmath716 .
we recall that for a subspace with fixed values of @xmath28 , @xmath26 , and @xmath27 , the number @xmath433 ( and @xmath717 ) is both the total number of @xmath13-spinons ( and spinons ) and the number of sites of the @xmath13-spin ( and spin ) effective lattice .
according to the studies of ref .
@xcite , the dimension of each such a subspace is , @xmath718 here the dimension @xmath719 and the two @xmath88 dimensions @xmath720 are given by , @xmath721 those are the number of @xmath52 fermion @xmath5 symmetry state representations and that of @xmath13-spin @xmath11 ( @xmath246 ) and spin @xmath11 ( @xmath215 ) symmetry state representations , respectively . the dimension @xmath719 given in eq .
( [ n - sa - ma ] ) is here straightforwardly recovered as @xmath722 .
it equals the number of occupancy configurations of the @xmath51 @xmath52 fermions in their @xmath52 band with @xmath233 discrete momentum values . on the other hand , the values of the numbers @xmath453 of discrete momentum values of the @xmath63 band
must exactly obey the following equality _ for all _ subspaces , @xmath723 here @xmath724 is the number of occupancy configurations of the set of @xmath462 @xmath63 fermions in their @xmath63 band with @xmath453 discrete momentum values .
the @xmath725 summation runs over _ the whole _ set of @xmath462 numbers that owing to the conservation of the @xmath88 numbers @xmath238 and @xmath239 of sites of the @xmath13-spin and spin effective lattices , respectively , whose occupancy configurations are not invariant under the electron - rotated - electron unitary transformation exactly obey the subspace sum rules given in eq .
( [ m - l - sum ] ) .
the general expression of the number @xmath453 of @xmath63 band discrete momentum values and thus of sites of the corresponding @xmath63 effective lattice that obeys eq . ( [ ncs - cpb ] ) for _ all _ subspaces is such that the @xmath392 ( and @xmath393 ) sites of the @xmath13-spin ( and spin ) effective lattice occupied by @xmath394 deconfined @xmath13-spinons ( and @xmath726 deconfined spinons ) and @xmath445 sites out of the @xmath418 sites of that lattice occupied by each local @xmath419 ( and @xmath420 ) fermion such that @xmath421 play the role of unoccupied sites of the @xmath60 ( and @xmath61 ) effective lattice .
indeed the unique solution of the problem that assures the validity of eq .
( [ ncs - cpb ] ) with the @xmath725 summation running over the whole set of @xmath462 numbers that obey the subspace sum - rule ( [ m - l - sum ] ) for _ all _ hilbert - space subspaces spanned by the states with fixed values of @xmath28 , @xmath26 , and @xmath27 refers to the following expression for the number of unoccupied sites of the @xmath63 effective lattice @xmath714 appearing in eq .
( [ n * ] ) , @xmath727 = [ n_{a_{\alpha}}^{d } - \sum_{\nu ' = 1}^{\infty}(\nu + \nu ' - \vert \nu-\nu'\vert)n_{\alpha\nu ' } ] \ , ; \hspace{0.25 cm } \alpha = \eta , s \ , ; \hspace{0.15 cm } \nu = 1,2, ... ,\infty \ , , \label{n - h - an}\ ] ] where @xmath728 .
importantly , this expression is also that obtained from the ba exact solution for the 1d hubbard model .
however , it is fully determined by state - representation dimension requirements of the global @xmath3 symmetry that apply to the hubbard model on the square lattice as well .
( note that the equivalence of the two expressions given in eq .
( [ n - h - an ] ) confirms that the @xmath51 and @xmath471 values with @xmath472 remaining fixed is equivalent to the @xmath13-spin @xmath26 ( @xmath246 ) or spin @xmath211 ( @xmath247 ) and values of the set of numbers @xmath471 with @xmath708 remaining fixed as well . in both cases that implies that the @xmath453 value remains fixed . )
the occupancies of the deconfined @xmath13-spinons and deconfined spinons give rise to the usual factors @xmath729 and @xmath730 , respectively , in the sub - space @xmath11 symmetry state - representation number expression @xmath731 and @xmath732 of eqs .
( [ n - sa - ma ] ) and ( [ ncs - cpb ] ) . in ref .
@xcite it is shown that the subspace - dimension summation , @xmath733}\,\sum_{s_{\eta}=0}^{[n_a^d/2-s_c]}\ , \sum_{s_s=0}^{s_c}{n_a^d\choose 2s_c } \prod_{\alpha = \eta , s}{[1+(-1)^{[2s_{\alpha}+2s_c]}]\over 2 } \,{\cal{n}}(s_{\alpha},m_{\alpha } ) \ , , \label{ntot}\ ] ] gives indeed the dimension @xmath44 of the full hilbert space . in the present context
this confirms that the set of momentum eigenstates @xmath107 of our description whose general expressions are given below in section iv - e is complete . from the use of the above equations one finds that the number of unoccupied sites of the @xmath734 effective lattices reads , @xmath735 \ , , \hspace{0.25 cm } \alpha = \eta \ , , s \
, , \label{nh+nh}\ ] ] where @xmath736 \ , , \hspace{0.15 cm } \alpha = \eta \ , , s \ , .
\label{sum - rules}\ ] ] the number @xmath737 given here equals that of @xmath734 fermion holes in the @xmath734 band . an important property discussed below in section iv - f is that for @xmath8 and @xmath601 the numbers @xmath738 and @xmath739 are conserved for the hubbard model on the square lattice in subspaces such that @xmath740 and @xmath741/n_a^2\rightarrow 0 $ ] as @xmath367 . for the 1d model they are good quantum numbers for the whole hilbert space .
( we could not prove that @xmath738 and @xmath739 are good quantum numbers for the hubbard model on the square lattice in its full hilbert space but that remains a possibility . )
straightforward manipulations of the above equations lead to the following general expressions for @xmath26 , @xmath27 , and @xmath28 , @xmath742 \ , , \hspace{0.25 cm } \alpha = \eta \ , , s \ , ; \hspace{0.50 cm } s_c = { 1\over 2}[n_a^d - n^h_{\eta 1 } - 2b_{\eta } ] = { 1\over 2 } [ n^h_{s1 } + 2b_{s } ] \ , .
\label{s - s - s}\ ] ] the equality of the two @xmath28 expressions given eq .
( [ s - s - s ] ) implies that , @xmath743 = [ n_a^d - \sum_{\alpha = \eta , s}2b_{\alpha } ] \ , .
\label{sum - nh+nh}\ ] ] provided that @xmath744 remains finite as @xmath196 , the related @xmath63 effective lattices can for the 1d and square - lattice models be represented by 1d and square lattices , respectively , of spacing , @xmath745 here @xmath746 and @xmath747 .
the arguments behind the lattice geometry and the average distance @xmath748 between the sites of the @xmath63 effective lattice playing the role of lattice spacing are similar to those used for the lattice geometry and spacing of the @xmath13-spin and spin effective lattices . in turn , the corresponding @xmath63 bands whose number of discrete momentum values is also given by @xmath453 are well defined even when @xmath453 is given by a finite small number , @xmath749 .
we start by introducing the lwss @xmath750 associated with the general momentum eigenstates @xmath751 , which refer to state representations of the model global @xmath3 symmetry . as justified in the following ,
the corresponding @xmath48 momentum eigenstates @xmath752 are of the form , @xmath753 [ \prod_{{\vec{q}}}{\mathcal{f}}^{\dag}_{{\vec{q}},c}\vert gs_c;0\rangle ] \ , .
\label{lws - full - el - infty}\ ] ] here @xmath754 and @xmath533 are creation operators of a @xmath48 @xmath63 fermion of momentum @xmath755 and @xmath52 fermion of momentum @xmath501 , respectively .
the @xmath13-spin @xmath11 vacuum @xmath255 containing @xmath263 deconfined @xmath58 @xmath13-spinons , the spin @xmath11 vacuum @xmath256 containing @xmath756 deconfined @xmath58 spinons , and the @xmath52 fermion @xmath5 vacuum @xmath259 appearing in this equation are those of the vacuum given in eq .
( [ vacuum ] ) .
( in its expression , @xmath757 . )
we recall that for @xmath48 eqs .
( [ fc+])-([c - up - c - down ] ) are equivalent to eqs .
( 1)-(3 ) of ref .
@xcite with the @xmath52 fermion creation operator @xmath162 replaced by the quasicharge annihilation operator @xmath185 .
therefore , the operator @xmath533 has the form given in eq .
( [ f+-cr ] ) .
the holes of such quasicharge particles are for @xmath48 the spinless and @xmath13-spinless @xmath52 fermions . in that limit
those describe the charge degrees of freedom of the electrons of the singly occupied sites .
moreover , the spinons and @xmath13-spinons of the @xmath59-spinon operators @xmath758 and @xmath59-@xmath13-spinon operators @xmath759 , respectively , are associated with the local spin and pseudospin operators , respectively , defined in that reference .
symmetry implies that for @xmath8 the @xmath453 @xmath63 band discrete momentum values of states belonging to the same @xmath103 tower are the same and thus @xmath10 independent .
importantly , the vacua of the @xmath48 states ( [ lws - full - el - infty ] ) are invariant under the electron - rotated - electron unitary transformation .
it follows that for finite values of @xmath10 the corresponding @xmath26 , @xmath199 , @xmath27 , @xmath200 , @xmath28 , and momentum eigenstates read , @xmath760 [ \prod_{{\vec{q}}}f^{\dag}_{{\vec{q}},c}\vert gs_c;0\rangle ] \ , , \nonumber \\
f^{\dag}_{{\vec{q}}\,',\alpha\nu } & = & { \hat{v}}^{\dag}\,{\mathcal{f}}^{\dag}_{{\vec{q}}\,',\alpha\nu}\,{\hat{v } } \ , ; \hspace{0.25 cm } f^{\dag}_{{\vec{q}},c } = { \hat{v}}^{\dag}\,{\mathcal{f}}^{\dag}_{{\vec{q}},c}\,{\hat{v } } \ , .
\label{lws - full - el}\end{aligned}\ ] ] unlike for the model on the square lattice , for the 1d model the states of form ( [ lws - full - el ] ) are energy eigenstates for the whole hilbert space @xcite . in appendix a it is confirmed that the discrete momentum values of the @xmath52 and @xmath63 fermion operators appearing on the right - hand side of eq .
( [ lws - full - el ] ) are indeed the good quantum numbers of the exact solution .
their occupancy configurations generate the energy eigenstates @xmath761 of general form ( [ lws - full - el ] ) .
the general state representations of the global @xmath3 symmetry associated with the momentum eigenstates @xmath107 are generated by simple @xmath52 fermion and @xmath63 fermion occupancy configurations and deconfined @xmath13-spinon and deconfined spinon occupancies of the form , @xmath762 [ \prod_{{\vec{q}}}f^{\dag}_{{\vec{q}},c}\vert gs_c;0\rangle ] \ , .
\label{non - lws}\end{aligned}\ ] ] here @xmath763 are the states of eq .
( [ lws - full - el ] ) and the normalization constant @xmath764 is given in eq .
( [ calpha ] ) of appendix b. the momentum values over which the products @xmath765 and @xmath766 , respectively , of the expressions provided in eqs .
( [ lws - full - el ] ) and ( [ non - lws ] ) run are those of the subspace of the state under consideration .
this justifies why the @xmath52 and @xmath63 bands have a set of suitable discrete momentum values so that the corresponding generators on the vacua are simple products of @xmath475 operators , as given in these equations .
the set of momentum eigenstates of form ( [ non - lws ] ) refers to an orthogonal and normalized state basis .
the number of such states equals the dimension given in eq .
( [ ntot ] ) . since in ref .
@xcite it is confirmed that such a dimension is @xmath44 , the set of states @xmath104 is complete .
consistently , for all values of the densities @xmath369 and @xmath43 the momentum eigenstates of general form ( [ non - lws ] ) correspond to the state representations of the @xmath3 group counted in section iv - d . also the related energy and momentum eigenstates @xmath272 considered below correspond to state representations of the @xmath3 group .
their @xmath13-spin - neutral @xmath13-spinon and spin - neutral spinon occupancy configurations that are not invariant under the electron - rotated - electron unitary transformation and involve @xmath238 and @xmath239 sites of the @xmath13-spin and spin effective lattices , respectively , are in general different from those of the momentum eigenstates ( [ non - lws ] ) .
the latter are simpler , referring to @xmath60 fermion and @xmath61 fermion occupancy configurations , respectively .
the unitary operator @xmath767 as defined in section ii has been constructed to inherently generating exact @xmath8 energy and momentum eigenstates @xmath768\vert \psi_{lws;u/4t}\rangle$ ] . within our description the corresponding set @xmath99 of @xmath48 energy and momentum eigenstates
are suitably chosen according to the recipe reported in such a section .
for the hubbard model on the square lattice at finite @xmath10 values the energy eigenstates @xmath272 are a suitable superposition of @xmath26 , @xmath199 , @xmath27 , @xmath200 , @xmath28 , @xmath542 , and momentum eigenstates @xmath107 ( [ non - lws ] ) of general form , @xmath769 [ \prod_{\alpha}\frac{({\hat{s}}^{\dag}_{\alpha})^{m^{de}_{\alpha,-1/2 } } } { \sqrt{{\cal{c}}_{\alpha}}}\sum_l c_l\prod_{\nu ( l)}\prod_{{\vec{q}}\,'(l ) } f^{\dag}_{{\vec{q}}\,',\alpha\nu}\vert 0_{\alpha};n_{a_{\alpha}^d}\rangle ] \ , .
\label{non - lws - psi}\end{aligned}\ ] ] here @xmath770 denote momentum eigenstates of form ( [ non - lws ] ) with different @xmath63 fermion momentum occupancy configurations labelled by the index @xmath771 also appearing in the products @xmath772 .
all such states have the same @xmath26 , @xmath199 , @xmath27 , @xmath200 , and @xmath28 values and the same momentum eigenvalue and momentum @xmath542 .
their @xmath52 fermion momentum distribution function @xmath773 is also the same , consistently with the form of expression ( [ non - lws - psi ] ) .
that for the hubbard model on the square lattice the states @xmath770 in the expansion ( [ non - lws - psi ] ) refer to different @xmath63 fermion momentum occupancy configurations follows from that model hamiltonian and the @xmath63 translation generators @xmath486 of eq .
( [ m - generators ] ) in general not commuting . although the numbers @xmath725 are not in general good quantum numbers for the hubbard model on the square lattice , the number @xmath217 of confined @xmath13-spinons @xmath242 and confined spinons @xmath243 is .
this means that rather being given by @xmath774 , such a number results from a different partition of the @xmath775 neutral confined @xmath13-spinon @xmath242 and confined spinon @xmath243 pairs . consistently , the numbers @xmath776 given in eq .
( [ n - h - an ] ) are not in general good quantum numbers of the hubbard model on the square lattice .
( for the 1d model they are . )
however , for a given energy eigenstate @xmath777 of general form provided in eq .
( [ non - lws - psi ] ) the values of the numbers of @xmath63 fermion holes @xmath778 of the corresponding momentum eigenstates @xmath770 are either even or odd integer numbers . indeed the numbers , @xmath779 are good quantum numbers of both the hubbard model on the square and 1d lattice . for the energy eigenstate under consideration
, the product @xmath780 appearing here runs over the same states as the sums @xmath781 in eq .
( [ non - lws - psi ] ) .
if @xmath778 is an even or odd integer for all such states , then @xmath782 is as well an even or odd integer , respectively . to arrive to the equality @xmath783 provided in eq .
( [ phs1 ] ) we use that @xmath784 $ ] , as given in eq .
( [ n - h - an ] ) , where @xmath785 is always an even integer number .
furthermore , from eq .
( [ m - l - sum ] ) one has that @xmath786 $ ] where @xmath787 is again always an even integer number .
moreover , @xmath788 where @xmath233 is an even integer and @xmath789 , as given in eq .
( [ m - l - c ] ) .
this justifies why @xmath790 and thus @xmath791 . in turn , from eq .
( [ l - l ] ) one has that @xmath792 where @xmath793 is an even integer , @xmath794 , and @xmath795 $ ] .
since the integer numbers @xmath271 and @xmath796 $ ] have the same parity , one finally finds that @xmath797 .
the number @xmath714 plays an active role in the physics provided that the corresponding @xmath63 fermion branch has finite occupancy @xmath798 at least for one of the states contributing to the sum @xmath781 of eq .
( [ non - lws - psi ] ) . for finite densities
@xmath799 this always holds for the number @xmath800 of @xmath64 fermion holes , which equals that of unoccupied sites of the @xmath64 effective lattice .
alike the set of momentum eigenstates @xmath104 of eq .
( [ non - lws ] ) , the complete set of energy and momentum eigenstates @xmath98 of general form given in eq .
( [ non - lws - psi ] ) refers to state representations of the group @xmath3 .
the only difference is that the @xmath238-sites @xmath13-spin - neutral and @xmath510-sites spin - neutral @xmath13-spinon and spinon configurations , respectively , that are not invariant under the electron - rotated - electron unitary transformation of the states @xmath98 are linear superpositions of those of the states @xmath104 . in some limiting cases such as for the energy eigenstates that span the one- and two - electron subspace the momentum - eigenstate superposition ( [ non - lws - psi ] ) reduces to a single state @xmath107 , so that @xmath801 . here
the index @xmath802 refers to the @xmath63 fermion occupancy configurations of the state @xmath107 of form ( [ non - lws ] ) under consideration .
( for 1d this applies to all states ( [ non - lws - psi ] ) . )
that @xmath26 , @xmath199 , @xmath27 , and @xmath200 are good quantum numbers for both the hubbard model on the 1d and square lattice implies that the numbers of deconfined @xmath13-spinons ( @xmath246 ) and deconfined spinons ( @xmath247 ) @xmath803 $ ] of eq .
( [ l - l ] ) are good quantum numbers as well .
the set of numbers @xmath804 of eq .
( [ phs1 ] ) and the @xmath17 momentum @xmath542 are also conserved for the model on both lattices . that the eigenvalue @xmath28 of the generator of the hidden @xmath5 symmetry of the model global @xmath3 symmetry is a good quantum number implies that the numbers @xmath805 and @xmath806 $ ] are also good quantum numbers .
this is why the numbers @xmath217 of eqs .
( [ 2s-2c ] ) , ( [ c - c ] ) , and ( [ m - l - sum ] ) , which can be expressed as @xmath807 $ ] , are good quantum numbers as well .
the same then applies to the total numbers @xmath808 of @xmath207 @xmath13-spinons ( @xmath246 ) and @xmath207 spinons ( @xmath247 ) .
the lack of integrability of the hubbard model on the square lattice is behind the @xmath63 translation generators @xmath486 of eq .
( [ m - generators ] ) not commuting with the hamiltonian and the set of numbers @xmath725 not being in general good quantum numbers .
however , @xmath738 and @xmath739 are for @xmath8 and @xmath601 good quantum numbers in subspaces such that @xmath740 and @xmath741/n_a^2\rightarrow 0 $ ] as @xmath367 . for the model in such subspaces that is a necessary condition for the @xmath17 momentum operator @xmath809 commuting with that model hamiltonian . that as reported above that commutator vanishes confirms that @xmath738 and @xmath739 are good quantum numbers in the above subspaces .
this is why they are called here quasi - good quantum numbers .
it follows from the relations ( [ nh+nh ] ) that the @xmath424 fermion hole and @xmath810 fermion hole numbers @xmath811 $ ] and @xmath812 $ ] , respectively , are conserved as well .
( it is possible that the numbers @xmath738 and @xmath739 and thus also the numbers @xmath813 and @xmath814 are conserved for the model on the square lattice in the whole hilbert space , yet we have not proved it . ) for some of the subspaces in which @xmath738 and @xmath739 have been shown to be good quantum numbers of the hubbard model on the square lattice most of the @xmath63 fermion numbers @xmath462 vanish and the remaining are good quantum numbers .
that happens for subspaces spanned by states with fixed values for the conserving numbers @xmath238 , @xmath239 , @xmath738 , and @xmath739 provided that such values are fulfilled by a unique choice for the corresponding values of the set of numbers @xmath725 .
this occurs for the following subspaces : \a ) subspaces for which @xmath815=0,1 $ ] for @xmath246 and/or @xmath247 . for such subspaces @xmath816
$ ] and @xmath817 $ ] are good quantum numbers and @xmath818 for @xmath819 . also @xmath820 $ ] is conserved .
furthermore , the @xmath48 local spin @xmath11 gauge symmetry for @xmath215 or the @xmath48 local @xmath13-spin @xmath11 gauge symmetry for @xmath214 also implies that the @xmath734 band momenta are good quantum numbers for the hubbard model on the square lattice in subspaces for which both the corresponding numbers @xmath821 and @xmath822 of unfilled and filled discrete momentum values , respectively , are conserved .
the unitarity of the operator @xmath89 then assures that in such subspaces the @xmath734 band momentum values are conserved for @xmath8 as well .
this is so for the present subspaces so that in them the hamiltonian of the hubbard model on the square lattice commutes with the @xmath734 translation generators @xmath823 and @xmath824 in the presence of the corresponding fictitious magnetic field @xmath825 .
\b ) a another type of subspace for which the @xmath63 fermion number @xmath412 is conserved is that for which @xmath826 and thus @xmath827
. then @xmath828 and @xmath829 for @xmath830 .
such subspaces have no no interest though for the problems studied in sections v and vi . for values of @xmath815 $ ] and @xmath831 other than those of the subspaces ( a ) and ( b )
there is no unique choice for the corresponding values of the set of numbers @xmath725 so that they are not good quantum numbers for the hubbard model on the square lattice .
indeed , states with the same @xmath815 $ ] and @xmath831 values and different values for the set of numbers @xmath725 may decay into each other . the general energy eigenstates ( [ non - lws - psi ] ) are superpositions of such states .
a particular case of a @xmath815=0 $ ] subspace refers to @xmath832 and @xmath833 $ ] . for it
@xmath834 $ ] , @xmath835 $ ] , and @xmath836 .
the equality @xmath837 ( and @xmath838 ) holds for all @xmath473 branches ( and @xmath573 branches ) .
for such a subspace @xmath839 ( and @xmath840 for @xmath573 ) are good quantum numbers . for @xmath841 , @xmath145 , and @xmath401 so that @xmath842 the hubbard model on a square - lattice in a @xmath843=0 $ ] subspace of type ( a ) has in the @xmath48 limit an interesting physics .
for such a subspace the @xmath64 fermion occupancy configurations are in that limit and within a mean - field approximation for the fictitious magnetic field @xmath615 of eq .
( [ a - j - s1 - 3d ] ) closely related to the physics of a full lowest landau level of the 2d quantum hall effect ( qhe ) . from the use of eqs . (
[ n * ] ) and ( [ n - h - an ] ) one finds @xmath844 .
that corresponds to the case in which @xmath845 .
if in the mean field approximation one replaces the corresponding fictitious magnetic field by the average field created by all @xmath64 fermions at a given position , one finds @xmath846\,{\vec{e}}_{x_3}$ ] . the number of @xmath64 band discrete momentum values
can then be written as , @xmath847 the bandwidth of the @xmath64 fermion energy @xmath848 vanishes in the @xmath48 limit .
hence only in that limit are the @xmath619 one-@xmath64-fermion states of eq .
( [ n - b - phi ] ) degenerate .
it follows that in such a limit @xmath621 plays the role of the number of degenerate states in each landau level of the 2d qhe . in the subspace considered here the @xmath64
fermion occupancies correspond to a full lowest landau level with filling factor one , as given in eq .
( [ n - b - phi ] ) . only for the @xmath48 limit
there is fully equivalence between the @xmath64 fermion occupancy configurations and , within a mean - field approximation for the fictitious magnetic field @xmath615 of eq .
( [ a - j - s1 - 3d ] ) , the 2d qhe with a @xmath849 full lowest landau level . in spite of the lack of state degeneracy emerging upon decreasing the value of @xmath10 ,
there remains for finite @xmath10 values some relation to the 2d qhe .
the occurrence of qhe - type behavior for the @xmath64 fermions in the square - lattice quantum liquid of ref .
@xcite shows that a magnetic field is not essential to the 2d qhe physics .
indeed , the fictitious magnetic field @xmath850 arises from expressing the effects of the electronic correlations in terms of the @xmath64 fermion interactions @xcite . the mean - field analysis associated with eq .
( [ n - b - phi ] ) is consistent with in the @xmath48 limit the hamiltonian of the hubbard model on the square lattice in a subspace ( a ) commuting with the @xmath64 translation generator @xmath557 .
indeed , in that limit such an analysis refers to an effective description where the hamiltonian is the sum of a @xmath52 fermion kinetic - energy term and a qhe like hamiltonian for the @xmath64 branch .
the @xmath64 translation generators @xmath557 then commute with all such hamiltonian terms .
this is consistent with in the @xmath48 limit its eigenvalues being good quantum numbers for the model on the square lattice in a subspace ( a ) . since the electron - rotated - electron transformation is unitary , such a commutation relation also holds for @xmath8 .
last but nor least , the states that span the one- and two - electron subspace introduced in the following are of type ( a ) .
this justifies why such states are energy eigenstates yet have the form ( [ non - lws ] ) .
hence for @xmath145 and excitation energy below @xmath402 for which @xmath401 , the square - lattice quantum - liquid momentum values of the @xmath52 , @xmath64 , and @xmath436 fermions are good quantum numbers @xcite .
for the half - filled hubbard model on the square lattice in the one- and two - electron subspace that holds as well for the @xmath424 band momentum values .
in this section we introduce a suitable one- and two - electron subspace .
( for hole concentrations @xmath145 our studies refer to excitation energies below @xmath402 for which @xmath401 for that subspace . ) for @xmath601 the picture that emerges is that of a two - component quantum liquid of charge @xmath52 fermions and spin neutral two - spinon @xmath64 fermions .
it refers to the square - lattice quantum liquid further investigated in ref .
@xcite .
we consider a @xmath601 and @xmath12 ground state @xmath852 whose @xmath52 and @xmath64 fermion occupancies are those given in section iv - c .
application onto it of a @xmath851-electron operator @xmath853 generates a state , @xmath854 contained in a @xmath851-electron subspace .
this is a subspace spanned by the set of energy eigenstates @xmath855 such that @xmath856 .
it is thus associated with a given @xmath851-electron operator @xmath857 .
generally , such operators can be written as a product of @xmath858 electron creation and annihilation operators . here
@xmath859 is the number of electron creation and annihilation operators in the operator @xmath857 expression for @xmath860 and @xmath861 , respectively , and with spin down and spin up for @xmath862 and @xmath863 , respectively .
a general local @xmath851-electron operator @xmath864 refers to a product of @xmath851 local electron creation and annihilation operators .
for @xmath865 such an operator has a well defined local structure .
it involves @xmath866 electron creation operators of spin projection @xmath867 and @xmath868 electron annihilation operators of spin projection @xmath867 whose real - space coordinates refer in general to a compact domain of neighboring lattice sites .
such a local @xmath851-electron operator @xmath864 may be labelled by the real - space coordinate @xmath869 of a corresponding central site .
a second type of @xmath851-electron operator is denoted by @xmath870 and carries momentum @xmath532 .
it is related to a local operator @xmath864 by a fourier transform .
the general @xmath851-electron operators @xmath857 considered here belong to one of these two types and are such that the ratio @xmath871 vanishes in the thermodynamic limit .
the operators @xmath870 of physical interest correspond in general to operators @xmath864 whose @xmath851 elementary electronic operators create or annihilate electrons in a compact domain of lattice sites .
the more usual cases for the description of experimental studies correspond to the @xmath872 one - electron and @xmath873 two - electron operators .
therefore , in this section we are mostly interested in the corresponding one- and two - electron subspace .
application onto a @xmath601 and @xmath566 ground state of a general @xmath851-electron operator @xmath857 leads to electron number deviations @xmath874 and @xmath875 . as a result of the expressions and relations of sections iii - b and iv - d
, such deviations may be expressed in terms of corresponding deviations in the number of @xmath52 fermions , @xmath63 fermions , deconfined @xmath13-spinons , and deconfined spinons as follows , @xmath876 and @xmath877 respectively .
( note that @xmath878 for the initial ground state so that @xmath879 , @xmath880 , @xmath881 , @xmath882 , and @xmath883 . )
hence only transitions to excited states associated with deviations obeying the sum rules ( [ dnu ] ) and ( [ dnd ] ) are permitted .
the electron number deviations ( [ dnu ] ) and ( [ dnd ] ) are associated with sum rules obeyed by the numbers @xmath859 of eq .
( [ d ] ) specific to the @xmath851-electron operator @xmath857 under consideration , which read , @xmath884 furthermore , it is straightforward to show that useful exact selection rules hold for excitations of well - defined initial ground states .
for instance , the values of the numbers @xmath885 = 2s_{\eta}$ ] of deconfined @xmath13-spinons and @xmath208 of @xmath207 deconfined @xmath13-spinons generated by application onto the @xmath886 ground state of a @xmath887-electron operator @xmath857 are restricted to the following ranges , @xmath888 respectively .
here @xmath26 denotes the excited - state @xmath13-spin and the numbers @xmath859 are those of eq .
( [ d ] ) specific to the @xmath887-electron operator . in the case on an initial @xmath841 ground state with hole concentration @xmath889 or @xmath890 one finds that the numbers @xmath891 = 2s_s$ ] and @xmath209 of deconfined spinons generated by application onto that state of a @xmath887-electron operator are restricted to the ranges , @xmath892 the range restrictions of eqs .
( [ lsrcsl ] ) and ( [ srcsl ] ) are exact for both the model on the square and 1d lattice , as well as for any other bipartite lattice .
for @xmath145 we limit our study to the vanishing rotated - electron double occupancy subspace considered in section iii - c . consistently with the @xmath338 energy spectrum of eqs .
( [ min - d ] ) and ( [ min - d - n ] ) , this is accomplished merely by limiting the excitation energy to values below @xmath402 , so that the @xmath401 constraint is automatically fulfilled : it follows from the form of such a spectrum that excited states with @xmath893 have vanishing rotated - electron double occupancy
. indeed creation of one rotated - electron doubly occupied site onto an initial @xmath145 and @xmath12 ground state is a process of minimum energy @xmath402 . in turn , at @xmath146 we consider both states with vanishing and finite rotated - electron double occupancy .
creation onto the @xmath894 , @xmath346 , and @xmath841 ground state of one @xmath60 fermion is a vanishing momentum process whose finite energy is exactly given by @xmath895 .
that object then obeys the criterion of eq .
( [ invariant - v ] ) for invariance under the electron - rotated - electron unitary transformation .
it follows that the @xmath13-spin degrees of freedom of such a @xmath60 fermion exactly involve @xmath473 electron doubly occupied sites .
furthermore , creation onto an initial @xmath601 and @xmath841 ground state of one @xmath61 fermion with a number @xmath573 of spinon pairs is a vanishing energy and momentum process . since vanishing spin
@xmath841 refers to vanishing magnetic field @xmath599 , such an object obeys the criterion @xmath896 of eq .
( [ invariant - v ] ) .
thus it is invariant under the electron - rotated - electron unitary transformation .
it follows that for @xmath8 creation of such an object involves occupancy configurations whose spin degrees of freedom are similar in terms of both rotated - electron and electron occupancy configurations .
that reveals that such a @xmath61 fermion describes the spin degrees of freedom of a number @xmath59 of electrons .
it then follows from the invariance under the electron - rotated - electron unitary transformation of the @xmath13-spin and spin degrees of freedom of the above @xmath60 fermion and @xmath61 fermion , respectively , that for @xmath889 where @xmath897 as @xmath196 and excitation energy @xmath898 and and any excitation energy at @xmath146 nearly the whole spectral weight generated by application onto the above ground states of @xmath887-electron operators refers to a subspace spanned by excited states with numbers in the following range , @xmath899 here @xmath774 where @xmath900 and @xmath901 , as given in eqs .
( [ m - l - sum ] ) and ( [ sum - rules ] ) , respectively .
the quantities @xmath902 , @xmath208 , @xmath238 , @xmath903 , @xmath209 , and @xmath510 appearing in eq .
( [ srs0 ] ) are good quantum numbers . moreover , according to the results of section iv - f , provided that @xmath904 and @xmath905/n_a^d\rightarrow 0 $ ] as @xmath906 , the number @xmath907 is a good quantum number for the model on the square lattice , alike for 1d .
the selection rules ( [ lsrcsl ] ) and ( [ srcsl ] ) are exact . in turn , for @xmath145 and @xmath12 initial ground states and excitation energy @xmath898 ( and an initial @xmath146 and @xmath12 ground state and any excitation energy ) nearly the whole @xmath887-electron spectral weight is generated by excited states whose numbers obey the approximate selection rules given in eq .
( [ srs0 ] ) .
indeed excited states with numbers @xmath908>{\cal{n}}$ ] for @xmath145 ( and @xmath909 and @xmath910>\sum_{l_s=\pm 1}{\cal{n}}_{+1,l_s}$ ] at @xmath146 ) generate a very small amount yet non vanishing @xmath887-electron spectral weight .
why in spite of the invariance under the electron - rotated - electron unitary transformation of the @xmath61 fermions with a number @xmath573 of spinon pairs created onto an initial @xmath601 and @xmath12 ground state ( and that of the @xmath60 fermions created onto an initial @xmath146 and @xmath12 ground state ) are the selection rules provided in eq .
( [ srs0 ] ) not exact ? the reason is that while the spin degrees of freedom of the @xmath59-electron occupancy configurations involved in a @xmath61 fermion are exactly described by that object , their hidden @xmath5 symmetry degrees of freedom are not invariant under the electron - rotated - electron unitary transformation .
the same applies to the @xmath13-spin degrees of freedom of the @xmath59-electron occupancy configurations exactly described by a @xmath60 fermion created onto a @xmath146 ground state .
their corresponding hidden @xmath5 symmetry degrees of freedom are not in general invariant under that transformation .
the spin degrees of freedom of the @xmath59-electron occupancy configurations involved in a @xmath61 fermion are for @xmath8 exactly the same as those of the corresponding @xmath59-rotated - electron occupancy configurations . in turn , the occupancy configurations of the @xmath59 @xmath52 fermions that describe the hidden @xmath5 symmetry degrees of freedom of the @xmath59 rotated electrons under consideration are slightly different from those of the corresponding @xmath59 electrons .
the same applies to the @xmath13-spin degrees of freedom and hidden @xmath5 symmetry degrees of freedom of the @xmath59 electrons and corresponding @xmath59 rotated electrons involved in a @xmath60 fermion created onto a @xmath146 ground state .
the former and the latter are and are not invariant under that transformation .
hence that the selection rules of eq .
( [ srs0 ] ) are not exact , yet are a very good approximation , stems from the lack of invariance under the electron - rotated - electron unitary transformation of the degrees of freedom associated with the hidden @xmath5 symmetry of the hubbard model .
this applies both to the model on the 1d and square lattice .
the @xmath887-electron spectral weight generated by excited states of initial @xmath145 and @xmath12 ground states of excitation energy @xmath898 and numbers @xmath910>{\cal{n}}$ ] is extremely small .
the same applies to excited states of initial @xmath146 and @xmath12 ground states of numbers @xmath909 and @xmath910>\sum_{l_s=\pm 1}{\cal{n}}_{+1,l_s}$ ] .
therefore , in this paper we define the @xmath851-electron subspace as that spanned by an initial @xmath601 and @xmath12 ground state plus the set of excited states whose numbers obey the approximate selection rules given in eq .
( [ srs0 ] ) .
note that the latter set of excited states depends on the specific @xmath887-electron operator under consideration .
( for hole concentrations @xmath145 this definition refers to excitation energy @xmath898 . )
the concept of a @xmath851-electron subspace as defined above refers to a specific operator .
in contrast , rather than referring to a specific @xmath851-electron operator , the one- and two subspace as defined here is the set of @xmath872 and @xmath873 subspaces associated with the one - electron operator and all simple two - electron operators , respectively . besides the @xmath872 one - electron operator @xmath911 ( measured in the angle - resolved photoelectron spectroscopy ) , this includes a set of @xmath873 operators @xmath870 such as the spin - projection @xmath912 density operator @xmath913\,\sum_{\vec{k}'}c^{\dagger}_{\vec{k}+\vec{k}',\sigma } c_{\vec{k}',\sigma}$ ] , the transverse spin - density operator @xmath914\,\sum_{\vec{k}'}c^{\dagger}_{\vec{k}+\vec{k}',\uparrow } c_{\vec{k}',\downarrow}$ ] , and the charge density operator ( measured in density - density electron energy loss spectroscopy and inelastic x - ray scattering ) .
the latter operator is written in terms of the above spin - up and spin - down density operators . moreover , the set of @xmath873 operators includes several superconductivity operators whose pairing symmetries are in general different at 1d and for the square lattice .
the local operators @xmath864 corresponding to the operators @xmath870 whose explicit expression is provided above read @xmath915 , @xmath916 , and @xmath917 , respectively . in the case of excitations of @xmath145 and @xmath12 ground states , in the remaining of this paper we are mostly interested in the subspace obtained from the overlap of the one- and two - electron subspace with the vanishing rotated - electron double occupancy subspace considered in section iii - c .
such a subspace is the one- and two - electron subspace for excitation energy @xmath898 . for finite hole concentrations
this it is the subspace of interest for the one- and two - electron physics . in turn , concerning the excitations of a @xmath146 and @xmath12 ground state we consider the whole one- and two - electron subspace , which refers both to the spin lower - hubbard band physics and one - electron and charge upper - hubbard band physics . as discussed above concerning the general @xmath887-electron spectral weight , for the model on the 1d and square lattices
there is for @xmath145 an extremely small amount of one- and two - electron spectral weight that for excitation energy @xmath898 is generated by states that do not obey the approximate selection rules of eq .
( [ srs0 ] ) for @xmath918 .
nearly all such very small amount of spectral weight refers to @xmath919 excited states .
( states with @xmath920 or @xmath921 generate nearly no spectral weight . )
that very small weight is neglected within our definition of the one- and two - electron subspace , which refers to excitation energies below @xmath402 . in turn , concerning the one- and two - electron excitations of a @xmath146 and @xmath12 ground state the very small amount of spectral weight generated by states that do not obey the approximate selection rules of eq .
( [ srs0 ] ) refers to @xmath919 and/or @xmath922 or @xmath923 excited states . that very small weight is also neglected within our definition of the one- and two - electron subspace .
( both the @xmath922 and @xmath923 excited states have energy much larger than the upper - hubbard band @xmath924 or @xmath925 excited states that obey the selection rules of eq .
( [ srs0 ] ) for @xmath918 . ) initial @xmath567 and @xmath12 ground states and their excited states of energy @xmath898 that span the one- and two - electron subspace considered in this paper have no @xmath57 @xmath13-spinons , no @xmath60 fermions , and no @xmath420 fermions with @xmath926 spinon pairs , so that @xmath927 and @xmath928 for @xmath926 . in turn ,
initial @xmath890 and @xmath12 ground states and their excited states that span the one- and two - electron subspace as defined here have no @xmath60 fermions with @xmath455 @xmath13-spinon pairs and no @xmath420 fermions with @xmath926 spinon pairs , so that @xmath927 for @xmath455 and @xmath928 for @xmath926 .
thus , consistently with the approximate selection rules of eq .
( [ srs0 ] ) , the values of the object numbers of such ground states and their excited states that span the subspace defined here are restricted to the following ranges , @xmath929 we emphasize that the maximum values of the numbers @xmath930 $ ] , @xmath931 $ ] , @xmath932 $ ] , and @xmath933 appearing in this equation must be consistent with the inequality requirement @xmath934 . furthermore , the hole concentrations in the inequality @xmath145 and equality @xmath146 also appearing here refer to the initial ground states and in the inequality @xmath145 we have neglected @xmath935 and @xmath936 corrections .
we recall that the numbers @xmath918 and @xmath859 correspond to a specific @xmath851-electron operator @xmath857 whose application onto a ground state @xmath852 generates @xmath851-electron excited states , as given in eq .
( [ 1 - 2-subspace ] ) .
the subspace defined here refers though to the whole set of such subspaces associated with the one - electron operator and the set of simple two - electron operators mentioned above .
fortunately , the subspace spanned by states whose numbers have values in the ranges given in eq .
( [ srs-0-ss ] ) is a @xmath843=n_{s2}=0,1 $ ] and @xmath937=0 $ ] subspace ( a ) , as defined in section iv - f . for initial @xmath145 and @xmath12 ( and @xmath146 and @xmath12 )
ground states it is such that @xmath938 $ ] , @xmath939 $ ] , and @xmath940 ( and @xmath938 $ ] , @xmath939 $ ] , and @xmath941 . ) hence the following @xmath52 and @xmath64 fermion numbers are conserved both for the subspaces with initial @xmath146 and @xmath145 ground states , @xmath942 = n_{a}^d \ , ; \hspace{0.25 cm } n_{c } = 2s_c \ , ; \hspace{0.25 cm } n^h_{c } = n_{a}^d - 2s_c \ , , \label{nac - nhc}\ ] ] and @xmath943 = [ s_c + s_s ] \ , ; \hspace{0.25 cm } n_{s1 } = [ s_c - s_s -2n_{s2 } ] \ , ; \hspace{0.25 cm } n^h_{s 1 } = [ 2s_s + 2n_{s2 } ] = 0,1,2 \ , , \label{nas1-nhs1}\ ] ] respectively . the hidden @xmath5 symmetry generator eigenvalue @xmath28 appearing here and the @xmath13-spin @xmath26 and spin @xmath27 have in the present subspace the following values , @xmath944 furthermore , for the subspace considered here the conserving number of eq .
( [ phs1 ] ) associated with the @xmath945 branch simplifies to , @xmath946 combination of this exact relation between @xmath800 and the number of electrons @xmath271 with the expressions and values of eqs .
( [ srs-0-ss ] ) and ( [ nas1-nhs1 ] ) reveals that the one- and two - electron subspace considered here is spanned by states whose deviation @xmath947 in the number of @xmath52 band holes and number @xmath539 of @xmath64 band holes may only have the following values , @xmath948 ( the initial @xmath601 and @xmath12 ground states have zero holes in the @xmath64 band so that @xmath949 for their excited states . ) as discussed below in section vi , for @xmath950 ground states and their charge excited states all @xmath49 spinons are confined within the two - spinon bonds of the @xmath951 @xmath64 fermions .
the number @xmath952 of eq .
( [ phs1 ] ) is associated with an important exact selection rule .
one of its consequences is that one- and two - electron excitations of @xmath601 and @xmath12 ground states contain no states with an even and odd number @xmath800 of @xmath64 fermion holes in the @xmath64 momentum band , respectively . for such initial ground states
we consider that @xmath271 is an even integer number .
since , as given in eq .
( [ phs1 ] ) , @xmath953 , for one - electron excited states for which the deviation in the value of @xmath271 reads @xmath954 the number of @xmath64 fermion holes @xmath800 must be always an odd integer . on the other hand , for both @xmath955 and @xmath956 two - electron excited states @xmath800 must be always an even integer .
such exact selection rules play an important role in the one- and two - electron spectra of the square - lattice quantum liquid further studied in ref .
@xcite . that quantum liquid refers to the hamiltonian ( [ h ] ) in the one- and two - electron subspace as defined here for initial @xmath601 and @xmath12 ground states .
for initial @xmath145 and @xmath12 ground states and excitation energy below @xmath402 the one- and two - electron subspace as defined here is spanned by the states of table 4 of ref .
@xcite . those are generated by creation or annihilation of @xmath957 holes in the @xmath52 momentum band and @xmath958 holes in the @xmath64 band plus small - momentum and low - energy particle - hole processes in the @xmath52 band .
the charge excitations of such initial ground states consist of a single particle - hole process in the @xmath52 band of arbitrary momentum and energy compatible with its momentum and energy bandwidths , plus small - momentum and low - energy @xmath52 fermion particle - hole processes .
such charge excitations correspond to state representations of the global @xmath5 symmetry and refer to the type of states denoted by charge " in the table 4 of ref .
the one - electron spin - doublet excitations correspond to the four types of states denoted by @xmath959el . " in that table where @xmath81 and @xmath82 denotes creation and annihilation , respectively , and @xmath960 .
the spin - singlet and spin - triplet excitations refer to the four types of states denoted by singl.spin " and tripl.spin " .
the two - electron excitations whose electrons are in a spin - singlet configuration and those whose two created or annihilated electrons are in a spin - triplet configuration correspond to the five types of states @xmath961el . " and @xmath962el . " where @xmath963 and @xmath964 denotes creation and annihilation , respectively , of two electrons .
the transformation laws under the electron - rotated - electron unitary transformation of deconfined spinons , deconfined @xmath13-spinons , and @xmath63 fermions of several branches were used above to show that nearly the whole one- and two - electron spectral weight of the excitations of the hubbard model on the 1d and square lattices is generated by processes obeying the ranges ( [ srs-0-ss ] ) and ( [ deltancs1 ] ) . upon expressing one- or two - electron operators @xmath853 in terms of rotated - electron creation and annihilation operators , they have in general an infinite number of terms , as given on the right - hand side of the first equation of ( [ oor ] ) .
such rotated - electron operator terms , which generate the excitations @xmath965 of eq .
( [ 1 - 2-subspace ] ) , may be expressed in terms of the @xmath52 fermion operators , spinon operators , and @xmath13-spinon operators given in eqs .
( [ fc+])-([rotated - quasi - spin ] ) .
this is done on using the operator relations provided in eq .
( [ c - up - c - down ] ) . concerning the contributions to the general operator expression given in eq .
( [ oor ] ) , which contains commutators involving the operator @xmath124 + { \cal{o } } ( t^2/u^2)$ ] , to fulfill such a task one takes into account that independently of their form , the additional higher - order operator terms @xmath966 are products of the kinetic operators @xmath125 , @xmath126 , and @xmath127 of eq . ( [ t - op ] ) . from such an analysis , one finds that the elementary processes associated with the one- and two - electron subspace number value ranges of eq .
( [ srs-0-ss ] ) are fully generated by the leading - order operator @xmath967 , which in our case is a one- or two - electron operator @xmath853 . in turn , the processes generated by the operator terms containing commutators involving the operator @xmath120 refer to excitations whose number value ranges are different from those provided in that equation .
this confirms that such processes generate very little one- and two - electron spectral weight , consistently with the exact number restrictions of eqs .
( [ lsrcsl ] ) and ( [ srcsl ] ) and the approximate number restrictions of eq .
( [ srs-0-ss ] ) . for the hubbard model on the 1d lattice
the spectral - weight distributions can be explicitly calculated by the pseudofermion dynamical theory associated with the model exact solution @xcite , exact diagonalization of small chains @xcite , and other methods . the relative one - electron spectral weight generated by different types of microscopic processes
is studied in ref .
the results of that reference confirm the dominance of the processes associated with the number value ranges provided in eq .
( [ srs-0-ss ] ) .
they refer specifically to operators @xmath968 and @xmath969 that are electron and rotated - electron , respectively , creation or annihilation operators .
such studies confirm that the operator @xmath967 generates _ all _ processes associated with the number value ranges of eq .
( [ srs-0-ss ] ) .
in addition , it also generates some of the non - dominant processes .
that is confirmed by the weights given in table 1 of ref .
@xcite , which correspond to the dominant processes associated with only these ranges .
the small missing weight refers to excitations whose number value ranges are not those of eq .
( [ srs-0-ss ] ) but whose weight is also generated by the operator @xmath967 .
indeed , that table refers to @xmath48 so that @xmath970 and the operator terms of the @xmath108 expression provided in eq .
( [ oor ] ) containing commutators involving the operator @xmath120 vanish . for finite values of @xmath10
all dominant processes associated with the number value ranges of eq .
( [ srs-0-ss ] ) are also generated by the operator @xmath967 . in turn , the small spectral weight associated with excitations whose number value ranges are different from those are generated both by that operator and the operator terms of the @xmath108 expression of eq .
( [ oor ] ) containing commutators involving the operator @xmath120 . for the model on the 1d lattice
the small one - electron spectral weight generated by the non - dominant processes is largest at half filling and @xmath971 .
the particle - hole symmetry of the @xmath146 and @xmath346 ground state implies that the relative spectral - weight contributions from different types of one - electron addition excitations given in fig . 2 of ref .
@xcite for the 1d model at half filling leads to similar corresponding relative weights for half - filling one - electron removal .
analysis of that figure confirms that for the corresponding one - electron removal spectrum the dominant processes associated with the number value ranges of eq .
( [ srs-0-ss ] ) refer to the states called holon - 1 @xmath64 hole states in figure 1 of ref .
@xcite . within our notation
such upper - hubbard band states have one deconfined @xmath58 @xmath13-spinon and one deconfined @xmath57 @xmath13-spinon for one - electron removal and addition , respectively .
their minimum relative weight of about @xmath972 is reached at @xmath971 . for other hole concentrations @xmath145 and values of @xmath10 the relative weight of one - electron states associated with the number value ranges of eq .
( [ srs-0-ss ] ) is always larger than @xmath972 , as confirmed from analysis of figs . 1 and 2 and the data provided in table 1 of that reference . for the hubbard model on the square lattice the explicit derivation of one- and two - electron spectral weights
is a more involved problem .
the number value ranges of eq .
( [ srs-0-ss ] ) also apply , implying similar results for the relative spectral weights of one- and two - electron excitations .
we emphasize that this is consistent with the exact range restrictions of eqs .
( [ lsrcsl ] ) and ( [ srcsl ] ) , the exact selection rule associated with the conservation of the number @xmath973 of eq .
( [ ns1h - general ] ) , and the transformation laws under the electron - rotated - electron unitary transformation of deconfined spinons , deconfined @xmath13-spinons , and @xmath63 fermions .
such transformation laws are behind the approximate number value ranges of eq .
( [ srs-0-ss ] ) being valid both for the hubbard model on the 1d and square lattices . according to the restrictions and numbers
values of eqs .
( [ srs-0-ss ] ) and ( [ deltancs1 ] ) , the states that span the one- and two - electron subspace may involve none or one @xmath436 fermion . as confirmed in the studies of ref .
@xcite , it is convenient to express the one- and two - electron excitation spectrum relative to initial @xmath601 and @xmath12 ground states in terms of the deviations in the numbers of @xmath52 effective lattice unoccupied sites and @xmath64 effective lattice unoccupied sites . those are given explicitly in eq .
( [ deltancs1 ] ) and equal the corresponding deviations in the numbers of @xmath52 band fermion holes and @xmath64 band fermion holes , respectively .
note that for @xmath145 and @xmath898 states the @xmath64 fermion related numbers provided in eq .
( [ nas1-nhs1 ] ) can be written as @xmath974 $ ] , @xmath975 $ ] , and @xmath976=0,1,2 $ ] where @xmath841 for @xmath977 and @xmath978 for @xmath979 . in turn , for excited states of the @xmath146 and @xmath12 ground state they read @xmath980 $ ] , @xmath981 $ ] , and @xmath976=0,1,2 $ ] where @xmath27 and @xmath982 may have the same values as above and @xmath983 for @xmath925 and @xmath984 for @xmath985 .
provided that @xmath369 is finite , the corrections to @xmath986 and @xmath987 are in both cases of the order of @xmath935 , whereas @xmath976=0,1,2 $ ] has the same expression and allowed values . as discussed above , for @xmath977 spin - singlet excited energy
eigenstates the single @xmath436 fermion has vanishing energy and momentum . consistently with eq .
( [ invariant - v ] ) , for vanishing magnetic field @xmath599 it is invariant under the electron - rotated - electron unitary transformation .
the same applies to the single @xmath424 fermion of @xmath925 @xmath13-spin - singlet excited states of the @xmath146 , @xmath346 , and @xmath12 ground state .
therefore , the only effect of creation and annihilation of such two objects is in the numbers of sites and occupied sites of the @xmath64 effective lattice .
their creation can then be merely accounted for by small changes in the occupancies of the discrete momentum values of the @xmath64 band .
hence and consistently with the additional information provided in appendix c , the only composite object whose internal occupancy configurations are important for the physics of the hamiltonian ( [ h ] ) in the one- and two - electron subspace is the spin - neutral two - spinon @xmath64 fermion and related spin - singlet two - spinon @xmath64 bond particle @xcite .
it turns out that for the hubbard model in the one- and two - electron subspace and alike for the @xmath436 fermion and/or the @xmath424 fermion , the presence of deconfined spinons is felt through the numbers of occupied and unoccupied sites of the @xmath64 effective lattice . for excited states of @xmath601 and @xmath12
ground states the number of deconfined @xmath13-spinons equals that of the unoccupied sites of the @xmath52 effective lattice . for excited states of the @xmath890 , @xmath988 , and @xmath12 ground state the presence of deconfined @xmath13-spinons
is felt in addition through the above numbers of sites and occupied sites of the @xmath64 effective lattice .
those may be rewritten as @xmath989 $ ] and @xmath990 $ ] , respectively , where @xmath991 .
therefore , when acting onto the one- and two - electron subspace as defined in this paper , the hubbard model refers to a two - component quantum liquid that can be described only in terms of @xmath52 fermions and @xmath64 fermions . for excited states of @xmath145 and
@xmath12 ground states this analysis applies to excitation energies @xmath898 .
for those of the @xmath146 , @xmath346 , and @xmath12 ground state the @xmath13-spin degrees of freedom are in addition behind the finite energy @xmath338 given in eq .
( [ min - d - n ] ) and the related energy @xmath992 . those are associated with rotated - electron doubly occupied sites and rotated - electron unoccupied sites , respectively , of excited states with finite occupancy @xmath993 of deconfined @xmath13-spinons or @xmath925 of a single @xmath424 fermion . however , the magnitude of that energy is fixed for each branch of excitations of that ground state .
hence for it and its excited states the square - lattice quantum liquid may again be described solely in terms of @xmath52 fermions and @xmath64 fermions . for excited states of @xmath145 and @xmath12 ground states belonging to the one- and two - electron subspace the spin effective lattice
has a number of sites given by @xmath994 .
for those of the @xmath146 , @xmath346 , and @xmath12 ground state it reads @xmath995 $ ] so that @xmath221 may have the values @xmath233 , @xmath996 $ ] , and @xmath997 $ ] . for @xmath145
its value is smaller than that of the original lattice . within the @xmath33 limit
one may neglect corrections of the order @xmath935 so that for both types of excited states the lattice spacing @xmath377 is that provided in eq .
( [ a - alpha ] ) for @xmath998 . for the model on the square lattice it reads @xmath999 , as given in eq .
( [ nncc ] ) .
both it and its general expression given in eq .
( [ a - alpha ] ) are such that the area @xmath1000 ^ 2=[a\times n_a]^2 $ ] of the system is preserved . as discussed in section iii - c ,
the concept of a spin effective lattice is valid only within the @xmath33 limit that our description refers to . for the model on the square lattice ,
the @xmath64 fermion spinon occupancy configurations considered in section vi are expected to be a good approximation provided that the ratio @xmath1001 and thus the electronic density @xmath369 remain finite as @xmath367 .
this is met for the hole - concentration range @xmath1002 considered in the studies of ref .
@xcite . within the present @xmath2 limit
there is for the one- and two - electron subspace of the model on the square lattice commensurability between the real - space distributions of the @xmath1003 sites of the @xmath64 effective lattice and the @xmath1004 sites of the spin effective lattice . for @xmath1005 and @xmath1006 the spin effective lattice has @xmath1007 sites and from the use of the expression given in eq .
( [ nas1-nhs1 ] ) for the number of @xmath64 effective lattice sites @xmath619 and eq . ( [ a - a - nu ] ) for the corresponding spacing @xmath1008 we find , @xmath1009 the general description introduced in this paper refers to a very large number of sites @xmath33 .
although very large , we assume that @xmath233 is finite and only in the end of any calculation take the @xmath196 limit . for @xmath33 very large but finite the @xmath12 ground state spin effective lattice is full and both at @xmath146 and for @xmath145 such a state is a spin - singlet state .
( for @xmath12 and @xmath146 this agrees with the exact theorem of ref .
@xcite . ) for @xmath1010 states such as such @xmath601 and @xmath12 ground states and their charge excited states the spin effective square lattice has two well - defined sub - lattices , which we call sub - lattice 1 and 2 , respectively .
( for the @xmath1011 states of the present subspace the spin effective lattice has two bipartite lattices as well , with the one or two extra sites accounted for by suitable boundary conditions . ) the two spin effective sub - lattices have spacing @xmath1012 .
the fundamental translation vectors of the sub - lattices 1 and 2 read , @xmath1013 respectively . here
@xmath1014 and @xmath1015 are unit vectors pointing in the direction associated with the cartesian coordinates @xmath1016 and @xmath1017 , respectively .
as further discussed in section vi , the vectors given in this equation are the fundamental translation vectors of the @xmath64 effective square lattice . in the case of @xmath601 , @xmath12 , and @xmath1010 ground states
whose @xmath64 momentum band is full and all @xmath1018 sites of the @xmath64 effective square lattice are occupied we consider that the square root @xmath365 of the number @xmath1019 of sites of the corresponding spin effective square lattice is an integer .
although the square root @xmath1020 of the number @xmath1018 of sites of the @xmath64 effective lattice is not in general an integer number , within the present @xmath2 limit we consider that it is the closest integer to it .
it follows from the the results reported in the previous sections and from the complementary technical analysis of appendix c that when acting onto the one- and two - electron subspace as defined in this paper , the hubbard model on a 1d or square lattice refers to a two - component quantum liquid described in terms of two types of objects on the corresponding effective lattices and momentum bands : the charge @xmath52 fermions and spin - neutral two - spinon @xmath64 fermions .
the one- and two - electron subspace can be divided into smaller subspaces that conserve @xmath28 and @xmath27 . when expressed in terms of @xmath52 and @xmath64 fermion operators , the hubbard model on a square lattice in the one- and two - electron subspace
is the square - lattice quantum liquid further studied in ref .
@xcite .
appendix c provides further information about why the square - lattice quantum liquid corresponding to the hubbard model on the square lattice in the one- and two - electron subspace may be described only by @xmath52 and @xmath64 fermions on their @xmath52 and @xmath64 effective lattices , respectively .
furthermore , the specific form that the momentum eigenstates of eq .
( [ non - lws ] ) have in the one- and two - electron subspace is provided in eqs .
( [ non - lws - c - s1 ] ) and ( [ lws - full - el - c - s1 ] ) of appendix c. consistently with the results of section iv - f concerning the subspaces of type ( a ) , for hubbard model on the square lattice in that subspace such states are energy eingenstates . however , we recall that the spin and @xmath64 effective lattices whose occupancy configurations generate the spin degrees of freedom of such states refer to an approximation valid only within the @xmath196 limit of our description . hence although the states of eqs .
( [ non - lws - c - s1 ] ) and ( [ lws - full - el - c - s1 ] ) of appendix c are exact energy eigenstates , our description refers to an approximate representation of such states .
since for @xmath145 and excitation energy @xmath1021 they span all subspaces of the one- and two - electron subspace that conserve @xmath28 and @xmath27 , they span the latter subspace as well .
we recall that states with a single @xmath436 fermion have also the general form provided eqs .
( [ non - lws - c - s1 ] ) and ( [ lws - full - el - c - s1 ] ) of appendix c. as discussed in section v - c and that appendix , the presence of that vanishing - energy , vanishing - momentum , and spin - neutral four - spinon object is accounted for the values of the numbers @xmath1022 $ ] and @xmath1023=0,1,2 $ ] of eq .
( [ nas1-nhs1 ] ) .
the quantum - liquid @xmath52 fermions are @xmath13-spinless and spinless objects without internal degrees of freedom and structure whose effective lattice is identical to the original lattice . for the complete set of @xmath8 energy eigenstates that span the full hilbert space , the occupied sites ( and unoccupied sites ) of the @xmath52 effective lattice correspond to those singly occupied ( and doubly occupied plus unoccupied ) by the rotated electrons .
the corresponding @xmath52 band has the same shape and momentum area as the first brillouin zone @xcite .
in contrast , the quantum - liquid composite spin - neutral two - spinon @xmath64 fermions have internal structure . thus the spinon occupancy configurations that describe such objects are for the one- and two - electron subspace a more complex problem discussed below in section vi .
it is simplified by the property that the number of unoccupied sites of the @xmath64 effective lattice is in that subspace limited to the values @xmath958 .
the energy eigenstates @xmath272 of general form given in eq .
( [ non - lws - c - s1 ] ) of appendix c that for @xmath145 and excitation energy @xmath1021 span the one- and two - electron subspace have numbers @xmath1024 and @xmath1025 such that @xmath1026=0,1,2 $ ] . as mentioned in section iv - f , for the model on the square lattice the spacing @xmath1027 of eq .
( [ a - a - s1-sube ] ) is directly related to a fictitious magnetic - field length @xmath1028 associated with the field of eq .
( [ a - j - s1 - 3d ] ) for the particular case of the @xmath616 branch . indeed , in that subspace one has that @xmath845 and such a fictitious magnetic field reads @xmath1029 .
it acting on one @xmath64 fermion differs from zero only at the positions of other @xmath64 fermions . in the mean - field approximation one replaces it by the average field created by all @xmath64 fermions at position @xmath160 .
this gives , @xmath1030\,{\vec{e}}_{x_3}=[\phi_0/a_{s1}^2]\,{\vec{e}}_{x_3}$ ] .
one then finds that the number @xmath619 of the @xmath64 band discrete momentum values equals @xmath1031/\phi_0 $ ] .
in addition , the @xmath64 effective lattice spacing @xmath1008 is expressed in terms to the fictitious magnetic - field length @xmath623 as @xmath1032 .
this is consistent with each @xmath64 fermion having a flux tube of one flux quantum on average attached to it . as further discussed in ref .
@xcite , for the present one- and two - electron subspace of the model on the square lattice the @xmath64 fermion problem is then related to the chern - simons theory @xcite .
indeed the number of flux quanta being one is consistent with the @xmath64 fermion and @xmath64 bond - particle wave functions obeying fermi and bose statistics , respectively .
hence the composite @xmath64 fermion consists of two spinons in a spin - singlet configuration plus an infinitely thin flux tube attached to it .
thus , each @xmath64 fermion appears to carry a fictitious magnetic solenoid with it as it moves around in the @xmath64 effective lattice . that the square - lattice quantum liquid is constructed to inherently the @xmath52 and @xmath64 fermion discrete momentum values being good quantum numbers is behind the suitability of the present description in terms of occupancy configurations of the @xmath52 and @xmath64 effective lattices and corresponding @xmath52 and @xmath64 band discrete momentum values .
the latter @xmath52 and @xmath64 values are the conjugate of the real - space coordinates of the @xmath52 and @xmath64 effective lattice , respectively .
are the approximations used in the construction of the @xmath64 effective lattice inconsistent with the @xmath64 band discrete momentum values being good quantum numbers ?
the answer is no . indeed , such approximations concern the relative positions of the @xmath1033 sites of the @xmath64 effective lattice @xcite .
those control the shape of the @xmath64 momentum band boundary .
they do not affect the @xmath64 band discrete momentum values being good quantum numbers . at @xmath146 the spin effective lattice
is identical to the original square lattice and the @xmath64 effective lattice is one of its two sub - lattices .
consistently , at @xmath146 and @xmath12 the boundary of the @xmath64 momentum band is accurately known .
then the @xmath64 band coincides with an antiferromagnetic reduced brillouin zone of momentum area @xmath1034 such that @xmath1035 @xcite . in turn
, it is known that for @xmath145 and @xmath12 the the @xmath64 band boundary encloses a smaller momentum area @xmath1036 yet its precise shape remains an open issue .
the related problems of the @xmath52 and @xmath64 momentum bands and corresponding energy dispersions and velocities are studied in ref .
in this section we call _ configuration states _ the spinon occupancy configurations in the spin effective lattice associated with the corresponding occupancy of the @xmath540 local @xmath64 fermions over the @xmath701 sites of the @xmath64 effective lattice .
a local @xmath64 fermion operator is generated from the corresponding @xmath64 bond particle operator @xmath1037 by the transformation @xmath1038 , as given in eq .
( [ f - an - operators ] ) for the @xmath616 branch .
here we consider the most general scenario according to which the @xmath64 bond particle is a superposition of all independent spin effective lattice two - site bonds centered at its real - space coordinate @xmath1039 .
this includes two - site bonds of all lengths , the only restriction being that they must be centered at the @xmath64 bond particle real - space coordinate @xmath1039 . within our general scheme ,
each two - site bond is associated with a coefficient whose exact magnitude remains unknown .
the exact 1d ba solution implicitly accounts for the magnitudes of such coefficients whereas for the model on the square lattice their exact magnitudes and bond and @xmath194 dependences remain open problems .
for instance , some such coefficients might vanish . depending on the values of the two - site - bond coefficients ,
our general formalism may describe both spin systems with different types of long - range or short - range orders and disordered spin systems .
the spin degrees of freedom of the 1d model is an example of a spin disordered system .
a @xmath601 and @xmath12 ground state and its charge excited states belonging to the one- and two - electron subspace are @xmath950 energy eigenstates whose spin degrees of freedom are described by the @xmath1010 configuration state studied in the following . within the use of suitable boundary conditions for the unoccupied site or two unoccupied sites of the @xmath1011 configuration states , similar results are obtained for such states . for @xmath1040 excited states , each configuration state refers to well - defined positions of the @xmath800 unoccupied sites .
the @xmath64 fermion operators @xmath1041 in the expression of the generators of the energy eigenstates whose general form is given in eq .
( [ non - lws - c - s1 ] ) of appendix c are a superposition of the local @xmath64 fermion operators provided in eq .
( [ f - an - operators ] ) for the @xmath616 branch .
the spin degrees of freedom of such states are generated by superpositions of the configuration states . while the states of eq .
( [ non - lws - c - s1 ] ) of appendix c refer to @xmath145 and excitation energy @xmath1021 , the same applies to the excited states of the @xmath146 , @xmath346 , and @xmath12 ground state that span the one- and two - electron subspace .
we consider torus periodic boundary conditions for the spin effective square lattice with @xmath1042 sites of the hubbard model on the square lattice , alike for the original lattice .
that implies periodic boundary conditions for the @xmath365 rows and @xmath365 columns .
periodic boundary conditions are used for the one - chain spin effective lattice of the 1d model .
the spin effective lattice has in the present case two sub - lattices , which we have named above sub - lattices 1 and 2 . for the model on the square lattice
their fundamental translation vectors are given in eq .
( [ a - b - s1 ] ) .
the real - space coordinates of the sites of each of such sub - lattices correspond to a possible choice of those of the @xmath64 effective lattice .
indeed , there is for the @xmath1010 configuration state a gauge symmetry " between the representations in terms of the occupancies of the two alternative choices of real - space coordinates of the @xmath64 effective lattice . in appendix b of ref .
@xcite it is confirmed that they refer to two alternative and equivalent representations of the @xmath1010 configuration state .
the possibility of the use of the real - space coordinates of either of the two corresponding choices of @xmath64 effective lattices to label the @xmath1010 configuration state is associated with the occurrence of a gauge structure @xcite .
the real - space coordinates @xmath1039 of such two sub - lattices have @xmath1043 sites , the fundamental translation vectors being those given in eq .
( [ a - b - s1 ] ) .
the real - space coordinates @xmath1039 of the local @xmath64 fermions and corresponding @xmath64 bond particles are chosen to correspond to those of one of these two sub - lattices . throughout the remaining of this section
the sub - lattices called sub - lattice 1 and sub - lattice 2 in section v - c are the sub - lattice of the spin effective lattice of a @xmath950 configuration state whose real - space coordinates are and are not the same as those of the @xmath64 effective lattice , respectively .
a two - site bond connects two sites of the spin effective lattice whose spinons have opposite spin projection and correspond to a two - site spin - singlet configuration defined below .
each @xmath64 bond particle is a suitable superposition of a well - defined set of two - site bonds .
such a spin - singlet two - spinon @xmath64 bond particle is related to the resonating - valence - bond pictures for spin - singlet occupancy configurations of ground states studied in refs .
the advantage of our rotated - electron description is that the @xmath64 bond particle is well defined for all values of @xmath8 and not only for @xmath206 .
this is consistent with its two spinons referring to spins of the sites singly occupied by rotated electrons rather than electrons .
indeed most schemes used previously for the hubbard model on the square lattice and related models involving singly - occupied - site spins refer in general to large values of @xmath206 @xcite . here
the @xmath64 bond particles have been constructed to inherently involving spinon occupancy configurations of sites of the spin effective lattice , which for @xmath8 refer only to the sites of the original lattice singly occupied by rotated electrons . for the @xmath1010 configuration state studied here the above bond superposition includes @xmath1044 families of two - site bonds , each family having @xmath1045 different types of such bonds : @xmath1045 is the largest number of independent two - site bonds with the same bond center that exist for the above - considered boundary conditions .
( above and in the remaining of this section we denote often the number of family two - site bonds by @xmath1044 where two and four is the number of such families for the model on the 1d and square lattice , respectively . )
two - site bond independence means here that for a given bond center all two - site bonds involve different pairs of sites and each site belongs to only one pair .
the set of independent two - site bonds with the same bond center belong to the same two - site bond family .
each two - site bond of a given family has some local @xmath64 fermion weight , which in some cases may vanish . for a @xmath64 bond particle of real - space coordinate @xmath1039
there are @xmath1044 families of two - site bonds .
the two - site bonds of each family are centered at one of the @xmath1044 points of real - space coordinate @xmath1046 . here
the indices @xmath1047 for @xmath37 , @xmath1048 for @xmath36 , and @xmath1049 uniquely define the two - site bond family and @xmath1050 is the primary link vector .
it connects the spin effective lattice site of real - space coordinate @xmath1039 to the center of the four ( and two for 1d ) two - site bonds of real - space coordinate @xmath1046 involving that site and its spin effective lattice nearest - neighboring sites .
note that the former site and the latter four sites ( and two sites for 1d ) belong to sub - lattice 1 and 2 , respectively .
on choosing one of the two sub - lattices of the spin effective lattice to be sub - lattice 1 and thus playing the role of @xmath64 effective lattice and representing the states in terms of the occupancy configurations of the latter lattice we say that there is a change of gauge structure @xcite . on accounting for all @xmath1045 independent two - site bonds for each of the @xmath618 two - site bond centers needed to describe a @xmath64 bond particle of real - space coordinate @xmath1039 we consider the most general situation .
the unknown exact configuration refers to some choice of the two - site bond local @xmath64 fermion weights considered below , some of which may vanish .
a @xmath64 bond particle of real - space coordinate @xmath1039 involves @xmath1051 two - site bonds .
this is consistent with each family having @xmath1052 two - site bonds of different type .
the two - site bond type is labeled by an index @xmath1053 $ ] uniquely defined below .
each two - site bond of a @xmath64 bond particle of real - space coordinate @xmath1039 involves two sites of coordinates @xmath1054 and @xmath1055 where @xmath1056 .
hence the two - site bond center @xmath1057 is the middle point located half - way between the two sites .
( the link vector @xmath1058 is defined below . )
the real - space coordinates @xmath1059 and @xmath1060 belong to the sub - lattice 1 and sub - lattice 2 of the spin effective lattice , respectively . for each family
there are @xmath1052 link vectors @xmath1058 , which read , @xmath1061 \ , , \label{r - r0-t}\ ] ] where the index @xmath1062 has values @xmath1047 , the index @xmath771 has values @xmath1063 , and @xmath1064 is a @xmath1065 vector .
it has cartesian components @xmath1066 $ ] for the square lattice and reads @xmath1067 $ ] for 1d .
there are @xmath1045 @xmath1065 vectors @xmath1064 , one for each choice of the following cartesian components , @xmath1068 here @xmath1069 , @xmath1070 , @xmath1071 , @xmath1072 and @xmath1073 and @xmath1074 are consecutive integer numbers .
the expressions provided in eq .
( [ r - r0-t ] ) apply to the 1d lattice provided that only the index value @xmath1048 is considered .
the single 1d component of the @xmath1065 vectors is given by @xmath1075 where @xmath1076 .
the two - site - bond - type index @xmath1053 $ ] labels the @xmath1045 @xmath1065 vectors @xmath1064 . for the model on the square lattice
it is defined in terms of the numbers @xmath1073 and @xmath1074 given in eq .
( [ xd - xd ] ) and reads , @xmath1077 for 1d one has that @xmath1078 .
the values of the two - site - bond - type index @xmath1079 are consecutive positive integers whose minimum value @xmath1080 corresponds to @xmath1081 so that , @xmath1082 for the model on the square lattice the maximum value @xmath1083 $ ] refers to @xmath1084 and @xmath1085 .
each pair of values ( and value ) of the cartesian coordinates of the @xmath1065 vector @xmath1066 $ ] for the square lattice ( and @xmath1067 $ ] for 1d ) corresponds to exactly one of the values @xmath1053 $ ] so that , @xmath1086 two - site bonds with the same @xmath1079 value and @xmath1087 and/or @xmath1088 are _ equivalent two - site bonds_. those are of the same type but belong to different families .
furthermore , @xmath1065 vectors @xmath1064 and @xmath1089 with the same value of @xmath1079 and @xmath1087 and/or @xmath1088 are related as follows , @xmath1090 { \vec{t}}_{d',l'}^{\,g } \ , ; \hspace{0.50 cm } { \bf \sigma_x } = \left [ \begin{array}{cc } 0 & 1 \\ 1 & 0 \nonumber \end{array}\right ] \ , , \label{t - t'-g}\ ] ] where @xmath1091 is the usual pauli matrix .
as described below in terms of suitable operators , a @xmath64 bond particle of real - space coordinate @xmath160 is a superposition of @xmath1092 two - site bonds , each being associated with a link vector @xmath1058 .
for each site of the spin effective lattice there is _ exactly one _ other site of the same lattice such that the bond connecting the two sites has center at @xmath1093 .
therefore , any two - site bond of the same family involves two sites of well - defined real - space coordinate @xmath1094 and @xmath1095 , which do not contribute together to any other two - site bond of the same family .
an important quantity is the distance between the two sites of a two - site bond , which we call _ two - site bond length_. it is independent of the real - space coordinate @xmath1096 of the two - site bond center and is fully determined by the link vector @xmath1058 and thus depends only on the two - site bond type associated with the index @xmath1079 . for @xmath37 and @xmath36 it reads , @xmath1097 respectively .
its minimum and maximum values are , @xmath1098 for the square lattice and @xmath1099 and @xmath1100 for 1d . for 1d ,
two - site bonds with different @xmath1079 values have different length @xmath1101 . in turn , for the square lattice there are two - site bonds of different type and hence different @xmath1079 values that have the same length @xmath1101 .
indeed , analysis of the two - site - bond - length expression of eq .
( [ xi - l ] ) reveals that for the latter lattice two - site bonds with different @xmath1079 values and numbers @xmath1102 $ ] and @xmath1103 $ ] , respectively , such that @xmath1104 and @xmath1105 have the same length .
the set of values of the numbers @xmath1106 and @xmath1107 given in eq .
( [ xd - xd ] ) for the model on the square lattice imply that the maximum value of the two - site bond length is @xmath1108 rather than @xmath1109 . indeed and as mentioned above , the two - site bonds contributing to a @xmath64 bond particle of the @xmath1010 configuration state are independent .
it is then required that each bond involves two sites that participate simultaneously in exactly one two - site bond . within the torus row and column periodic boundary conditions for the spin effective square lattice
such a requirement is fulfilled provided that the range of the numbers @xmath1106 and @xmath1107 is that given in eq .
( [ xd - xd ] ) .
for each family of two - site bonds associated with a @xmath64 bond particle of real - space coordinate @xmath1110 there is a _ primary two - site bond_. it corresponds to @xmath1080 and thus connects two nearest - neighboring sites of the spin effective lattice , one of them having the same real - space coordinate @xmath1111 as the @xmath64 bond particle and corresponding local @xmath64 fermion . for primary two - site bonds
the link vector @xmath1058 reads @xmath1112 where the primary link vector @xmath1050 is given in eq .
( [ r - r0-t ] ) . for the model on the square lattice
there are four primary two - site bonds , one per family .
their link vectors @xmath1050 have components such that @xmath1081 in eqs .
( [ r - r0-t ] ) and ( [ xd - xd ] ) .
therefore , the primary two - site bonds have minimum length @xmath1113 . alike
the remaining two - site bonds of its family , the center of a primary two - site bond is located at @xmath1096 .
for the square lattice there are two horizontal primary two - site bonds whose centers are located at @xmath1114 with @xmath1049 and two vertical primary two - site bonds whose centers are located at @xmath1115 with @xmath1049 . in the case of the 1d lattice
there are two primary two - site bonds whose centers are located at @xmath1114 with @xmath1049 .
the building blocks of a @xmath950 configuration state are singlet pairs of spinons on sites @xmath1116 and @xmath1117 of the spin effective lattice , @xmath1118 here , as given above , the values of the integer indices @xmath1062 , @xmath771 , and @xmath1079 are in the ranges @xmath1047 , @xmath1049 , and @xmath1119 , respectively .
such values are a function of the index @xmath1043 , where @xmath1120 , of the real - space coordinate @xmath160 of each @xmath64 bond particle in the sub - lattice 1 .
indeed , the two sites of such pairs of sites are connected by two - site bonds and each bond is associated with exactly one @xmath64 bond particle .
each connection involving @xmath1121 different bonds determines a _
partition_. a partition is a @xmath221-spinon occupancy configuration where each site of the spin effective lattice is linked to one site only and all @xmath1122 sites then correspond to @xmath1121 well - defined two - site bonds , each belonging to a different @xmath64 bond particle .
each of the @xmath540 @xmath64 bond particles contributes with exactly one of its @xmath1051 two - site bonds to a partition . in a partition
any site of the spin effective lattice participates in one bond only and there is a single two - site bond attached to each site , which connects it to some other site . and the latter site is attached to the former site only .
the @xmath950 configuration state may be represented as , @xmath1123 where the coefficients @xmath1124 are associated with the local @xmath64 fermion bond weights and appear in the local @xmath64 fermion operators defined below , @xmath1125 is given in eq .
( [ f - an - operators ] ) for the @xmath945 branch , the product of singlet states @xmath1126 refers to a bond state associated with a given partition , and the summation @xmath291 runs over all partitions .
a particular type of partition involves @xmath1127 identical two - site bonds . the indices @xmath1062 , @xmath771 , and @xmath1079 of identical two - site bonds have the same values but correspond to @xmath64 bond particles with different real - space coordinates @xmath160 .
such a partition involves a set of @xmath1127 identical two - site bonds whose bonds connect different sites of the spin effective lattice , each site being linked to exactly one site . in this case the two real - space coordinates of the @xmath1127 pairs of sites are connected by the same real - space vector @xmath1128 , so that each two - site bond has the same length .
horizontal two - site bonds for the model on the square lattice .
the primary two - site bonds are represented by the thick horizontal lines connecting two sites of the spin effective lattice . for a reference frame where the site located at the corner on the left - hand side and lower limit of the squared sub - domain has cartesian coordinates @xmath1129 the family indices read @xmath1048 , @xmath1130
if that site belongs to sub - lattice 2 , whereas @xmath1048 , @xmath1131 if instead it belongs to sub - lattice 1.,width=132,height=132 ] each of the @xmath1051 bonds of a partition involves two sites of real - space coordinates @xmath1132 and @xmath1133 that belong to different sub - lattices , where @xmath1134 .
the relation to the notation used above for the real - space coordinates of the two sites of a bond is as follows , @xmath1135 here both @xmath1132 and @xmath1110 are real - space coordinates of the sub - lattice 1 and thus of the @xmath64 effective lattice . except for a primary two - site bond , one has that @xmath1136 .
the site of real - space coordinate @xmath1110 is that of the corresponding local @xmath64 fermion .
it is the closest to the two - site bond center at @xmath1093 . in turn
, @xmath1132 is the real - space coordinate of one of the two sites of the spin effective lattice involved in the two - site bond . when a partition is a set of @xmath1051 identical primary two - site bonds , all site pairs involve nearest - neighboring sites of the spin effective lattice .
it is then called a _ primary partition_. the family of a primary partition
is labeled by the indices @xmath1062 and @xmath771 of the corresponding identical two - site bonds .
figures 1 and 2 represent primary partitions of @xmath1048 horizontal and @xmath1137 vertical two - site bonds , respectively , for a sub - domain of the spin effective lattice of the model on the square lattice .
a useful concept is that of a _
@xmath1079-primary partition_. it is defined as the superposition of the @xmath1044 primary partitions .
it follows that a @xmath1079-primary partition contains @xmath1138 primary two - site bonds .
in such a configuration each of the @xmath1122 sites of the spin - effective lattice has @xmath1044 two - site bonds attached to it .
figure 3 shows a sub - domain of the spin - effective lattice with the @xmath1079-primary partition of the @xmath1010 configuration state for the model on the square lattice .
the spinon occupancy configurations considered above are similar to those associated with multi - spin wave functions of spin - singlet states used by several authors @xcite .
such wave functions are often constructed having as building blocks two - site and two - spin spin - singlet configurations similar to that of eq .
( [ state-+ ] ) except that here the two corresponding spinons refer to sites singly occupied by rotated electrons .
thus here they correspond to @xmath8 rather than only to @xmath206 for the previous approaches . in those one also connects pairs of lattice sites with bonds and
each such a connection determines a partition .
however , here bonds involve sites of the spin effective lattice whereas those of previous related studies refer to the sites of the original lattice of the corresponding quantum problems . for a given partition one can define a valence bond state @xcite as a product of singlet states and represent an arbitrary singlet by a superposition of valence bond states of general form similar to that of eq .
( [ phi-+ ] ) . that involves a sum over all partitions of the lattice into set of pairs .
however , for general wave functions such a decomposition works in general very badly .
indeed , valence - bond states are not orthogonal and their basis is overcomplete .
fortunately , here each of the @xmath1127 two - site bonds of a partition belongs to a different @xmath64 bond particle so that each of such particles contributes to a partition with exactly one bond .
such a restriction eliminates the unwanted and unphysical contributions and renders the bond states free of the overcompleteness problem .
the @xmath64 bond particle operators @xmath1139 and @xmath1140 may be expressed in terms of superpositions of two - site bond operators .
we recall that according to eq .
( [ f - an - operators ] ) for the @xmath616 branch , the operator @xmath1037 is related to the corresponding local @xmath64 fermion operator by the transformation @xmath1038 . for the one- and two - electron subspace spanned by states whose numbers obey the approximate selection rules of eq .
( [ srs-0-ss ] ) as @xmath196 , the operators @xmath1139 ( and @xmath1037 ) that annihilate ( and create ) a @xmath64 bond particle at a site of the spin effective lattice of real - space coordinate @xmath1039 have the following general form both for the @xmath1141 and square lattices , @xmath1142 the operators @xmath1143 and @xmath1144 appearing in these expressions create and annihilate , respectively , a superposition of @xmath1044 two - site bonds of the same type and @xmath1145 and @xmath1146 are two - site bond operators whose expression is given below .
vertical two - site bonds for the model on the square lattice .
the primary two - site bonds are represented by the thick vertical lines connecting two sites of the spin effective lattice . alike in fig .
1 , the family indices read @xmath1137 , @xmath1130 if the site of cartesian coordinates @xmath1129 belongs to sub - lattice 2 , whereas @xmath1137 , @xmath1131 if instead it belongs to sub - lattice 1.,width=132,height=132 ] for the model on the square ( and 1d ) lattice the four ( and two ) primary two - site bonds associated with the operators @xmath1147 and @xmath1148 are behind most of the spectral weight of a local @xmath64 fermion and corresponding @xmath64 bond particle of real - space coordinate @xmath1110 . consistently , the absolute value @xmath1149 of the coefficients @xmath1150 appearing in the expressions of such operators given in eq .
( [ g - s1+general ] ) decreases for increasing two - site bond length @xmath1101 .
these coefficients obey the normalization sum - rule , @xmath1151 } \vert h_{g}\vert^2 = { 1\over 2d } \ , ; \hspace{0.50 cm } d=1,2 \ , .
\label{g - s1+sum - rule}\ ] ] the exact dependence of @xmath1149 on the length @xmath1101 , value of @xmath10 , and hole concentration @xmath194 remains for the hubbard model an involved open problem .
the suitable use of this sum - rule and related symmetries leads though to useful information , as discussed below .
( the real - space coordinates @xmath1039 of the local @xmath64 fermion operators and corresponding @xmath64 bond - particle operators of eq .
( [ g - s1+general ] ) have been chosen to inherently being those of the sub - lattice 1 . )
the two - site bond operators @xmath1145 and @xmath1146 appearing in eq .
( [ g - s1+general ] ) are associated with a well - defined bond connecting the two sites of real - space coordinates @xmath1054 and @xmath1060 , respectively .
their expression can be obtained by considering the following related operator , @xmath1152 \nonumber \\ & = & f_{\vec{r}-\vec{r}_{d , l}^{\,g},c}^{\dag}\,f_{\vec{r}+\vec{r}_{d , l}^{\,g},c}^{\dag } \ , b_{\vec{r},s1,d , l , g}^{\dag } \ , .
\label{singlet - conf}\end{aligned}\ ] ] its first expression provided here is in terms of the rotated - electron operators of eq .
( [ rotated - operators ] ) .
the second expression is in terms of the @xmath52 fermion operators given in eq .
( [ fc+ ] ) and the following two - site bond operator , @xmath1153 s^-_{\vec{r}+\vec{r}_{d , l}^{\,g } } - \left[{1\over 2}+s^{x_3}_{\vec{r}+\vec{r}_{d , l}^{\,g}}\right ] s^-_{\vec{r}-\vec{r}_{d , l}^{\,g}}\right ) \ , , \label{g - s - l}\ ] ] such that @xmath1154 . here
the spinon operators are those given in eq .
( [ sir - pir ] ) . the second expression of eq . ( [ singlet - conf ] ) is obtained from the use of eq .
( [ c - up - c - down ] ) .
-primary partition of a @xmath1010 configuration state for the model on the square lattice .
the sites belonging to the sub - lattice 1 are represented by filled circles .
the horizontal ( and vertical ) thin and thick lines refer to @xmath1048 ( and @xmath1137 ) @xmath1130 and @xmath1131 primary two - site bonds , respectively . in a @xmath1079-primary partition of the model on the square lattice each site of the spin effective lattice
has four two - site bonds attached to it.,width=141,height=170 ] on combining the local @xmath64 fermion operator @xmath1155 expression given in eq .
( [ f - an - operators ] ) for the @xmath945 branch with eqs .
( [ g - s1+general ] ) and ( [ g - s - l ] ) one arrives to the following equations for the operators @xmath1155 and @xmath1156 , @xmath1157 s^-_{\vec{r}_{j}+{\vec{r}_{d , l}}^{\,0}+\vec{r}_{d , l}^{\,g } } - \left[{1\over 2}+s^{x_3}_{\vec{r}_{j}+{\vec{r}_{d , l}}^{\,0}+\vec{r}_{d , l}^{\,g}}\right ] s^-_{\vec{r}_{j}+{\vec{r}_{d , l}}^{\,0}-\vec{r}_{d , l}^{\,g}}\right ) \ , , \nonumber \\ \phi_{j , s1 } & = & \sum_{j'\neq j}f^{\dag}_{{\vec{r}}_{j'},s1 } f_{{\vec{r}}_{j'},s1}\,\phi_{j',j , s1 } \ , ; \hspace{0.35 cm } \phi_{j',j , s1 } = \arctan \left({{x_{j'}}_2-{x_{j}}_2\over { x_{j'}}_1-{x_{j}}_1}\right ) \ , .
\label{f - s1-operator}\end{aligned}\ ] ] the phase factor @xmath1158 that appears both here and in the operator of eq .
( [ g - s - l ] ) is associated with the @xmath1062-wave symmetry of the @xmath64 fermion spinon pairing of the model on the square lattice .
the introduction of such a phase - factor refers to a self - consistent procedure .
it follows from the @xmath1062-wave symmetry of the energy dispersion found in ref .
@xcite for the @xmath64 fermions .
their energy - dispersion @xmath1062-wave symmetry arises naturally from symmetries beyond the form of the operators introduced in eqs .
( [ singlet - conf ] ) and ( [ g - s - l ] ) . according to the above discussions ,
the real - space coordinates @xmath1054 and @xmath1060 involved in the operators of eqs .
( [ singlet - conf ] ) and ( [ g - s - l ] ) and in the operators of eq .
( [ f - s1-operator ] ) for @xmath1096 correspond to two sites that belong to different sub - lattices of the spin effective lattice .
moreover , @xmath1096 is the two - site bond center and the primary link vector @xmath1050 and link vector @xmath1058 are given in eqs .
( [ r - r0-t ] ) and ( [ xd - xd ] ) , respectively . in the configuration generated by the operator of eq . ( [ singlet - conf ] ) the two sites
are singly occupied by the rotated electrons associated with the operators appearing in the first expression of that equation .
the @xmath1010 configuration state ( [ phi-+ ] ) can be written in terms of @xmath64 fermion operators @xmath1155 defined by eq .
( [ f - s1-operator ] ) as follows , @xmath1159 here @xmath256 is the spin @xmath11 vacuum on the right - hand side of eq .
( [ vacuum ] ) whose number @xmath265 of deconfined @xmath58 spinons equals in the present case the number @xmath1160 of confined spinons contained in the whole set of @xmath540 @xmath64 fermions of the state @xmath1161 under consideration .
such a vacuum corresponds to a fully polarized spin - up spinon configuration . on imposing the equality of the general configuration states given in eqs .
( [ phi-+ ] ) and ( [ phi-+-s1 ] ) , respectively , one defines which of the processes generated by application of the operators of eqs .
( [ g - s1+general])-([f - s1-operator ] ) onto the neutral states that span the subspaces of the one- and two - electron subspace are physical .
indeed the equality of the general configuration states given in such two equations implies several restrictions onto the processes generated by the two - site bond operators ( [ g - s - l ] ) .
those can be summarized in a few corresponding rules for exclusion of the unwanted spin configurations generated by unphysical processes . in order to introduce such rules ,
it is useful to briefly report and discuss a few issues related to the local @xmath64 fermion operators and corresponding @xmath64 bond - particle operators .
those follow from the algebra given in eqs .
( [ albegra - s - p - m])-([albegra - s - sz - com ] ) of appendix d for the basic spinon operators of the two - site bond operators expressions provided in eq .
( [ g - s - l ] ) . the @xmath64 bond - particle operators @xmath1037 and @xmath1139 of eq .
( [ g - s1+general ] ) and corresponding local @xmath64 fermion operators @xmath1155 and @xmath1156 of eq .
( [ f - s1-operator ] ) involve a sum of @xmath1127 two - site bond operators of general form given in eq .
( [ g - s - l ] ) with @xmath1162 of such operators per family . the number of unoccupied sites @xmath814 of eq .
( [ nas1-nhs1 ] ) refers to a subspace with a fixed number @xmath540 of @xmath64 fermions . in turn , the creation and annihilation of one local @xmath64 fermion by application of the operators @xmath1155 and @xmath1156 , respectively , onto the ground state involves a superposition of @xmath1127 elementary processes , which do not conserve the number of these objects .
each such an elementary process is generated by an operator @xmath1145 and @xmath1146 , respectively , whose expression is given in eq .
( [ g - s - l ] ) . within the lws representation
mostly used in this paper , application of the rotated - electron operators of eq .
( [ singlet - conf ] ) onto two rotated - electron unoccupied sites of the original lattice generates two virtual processes .
the first process involves creation of two @xmath52 fermions and two deconfined @xmath58 spinons .
the second process refers to creation of a spin - singlet two - site and two - spinon configuration upon annihilation of two deconfined @xmath58 spinons .
indeed , from analysis of the expression provided in eq .
( [ g - s - l ] ) , one finds that application of the operator @xmath1145 onto the sites of real - space coordinates @xmath1054 and @xmath1055 gives zero except when both these sites refer to a deconfined @xmath58 spinon .
this is consistent with an `` occupied site '' of the @xmath64 effective lattice corresponding to two sites of the spin effective lattice with real - space coordinates @xmath1054 and @xmath1060 , respectively , which refer to deconfined @xmath58 spinons in the initial configuration state . consistently with the studies of ref .
@xcite , for excitations involving transitions between configuration states under which a @xmath64 fermion moves around in the @xmath64 effective lattice through elementary processes that conserve the numbers @xmath540 and @xmath800 , a deconfined @xmath58 spinon plays the role of a unoccupied site of such an effective lattice .
in contrast , in elementary processes involving the creation of a @xmath64 fermion the two annihilated deconfined @xmath58 spinons give rise to a @xmath64 effective lattice `` occupied site '' in the final occupancy configuration .
-primary partitions associated with @xmath1010 and @xmath1163 configuration states for the model on the square lattice . in the configuration of the figure on the left - hand side the filled circle at the middle of the sub - domain corresponds to the site
whose real - space coordinate @xmath160 is that of the @xmath64 fermion .
annihilation of that object leads to the configuration on the figure right - hand side .
note that for a primary @xmath1079-basic partition this is equivalent to the suppression of the four two - site bonds attached to the above site of real - space coordinate @xmath160.,width=190,height=68 ] according to the operator expression provided in eq .
( [ g - s - l ] ) , upon acting onto the deconfined - spinon occupancies the operator @xmath1145 generates a superposition of two configurations . for one of those the elementary process generated by that operator flips the spin of the spinon at site @xmath1060 and checks
whether the spin of the spinon at site @xmath1054 remains up .
the elementary process generating the other configuration flips the spin of the spinon at site @xmath1054 and checks whether the spin of the spinon at site @xmath1060 remains up .
the relative phase factor @xmath82 of the two configurations insures that the @xmath64 fermion created by the operator @xmath1164 of eq .
( [ f - s1-operator ] ) is a suitable superposition of spin - singlet configurations .
figure 4 shows a small sub - domain of two @xmath1079-primary partitions of @xmath1010 and @xmath1163 configuration states , respectively . in the configuration on the figure left - hand side the filled circle at the middle of the sub - domain corresponds to a site
whose real - space coordinate @xmath160 is that of a local @xmath64 fermion .
application of the annihilation operator @xmath1165 of eq .
( [ f - s1-operator ] ) onto that @xmath1079-primary partition occurs through the operator @xmath1148 also given in that equation .
that leads to the configuration on the figure right - hand side .
hence annihilation of the local @xmath64 fermion is for a @xmath1079-primary partition equivalent to the suppression of the four two - site bonds attached to the above site of real - space coordinate @xmath160 . a superficial analysis of the configurations shown in fig .
4 seems to indicate that there is one unoccupied site in the sub - domain of the final @xmath1163 configuration state .
however , if instead of the @xmath1079-primary partition one considers the corresponding four primary partitions , one finds that there are two nearest - neighboring unoccupied sites . for the @xmath1048 and @xmath1137 primary partitions these two sites belong to the same row and column , respectively . concerning the application of two - site bond operators and products of two - site bond operators onto spin configurations the restrictions arising from imposing that the representations of the @xmath950 configuration state given in eqs .
( [ phi-+ ] ) and ( [ phi-+-s1 ] ) , respectively , are identical correspond to the following simple three rules whose fulfillment prevents the generation of unwanted and unphysical spin configurations : _ first rule _
according to which application onto a spin configuration of a two - site bond operator generates a physical spin configuration provided that in the initial spin configuration its sites of real - space coordinates @xmath1054 and @xmath1060 , respectively , ( i ) refer to deconfined @xmath58 spinons or ( ii ) are linked by a bond .
_ second rule _ states that in the processes generated by application onto a spin configuration of a product of a set of two - site bond operators the first rule applies to each operator provided that ( i ) the two sites of each operator are not joined by any other of the two - site bond operators of the set or ( ii ) both sites of some or all such operators are the same as the two sites of one or several other operators . otherwise , application onto a spin configuration of the set of two - site bond operators gives zero .
hence , each two - site bond operator can not join a single site with another operator of the set : it either joins none or both sites with one or several such operators . _
_ refers to when the two - site bond operators of a product of a set of such operators correspond to the same @xmath64 bond particle and the first and second rules are obeyed .
then the processes generated by application onto a spin configuration of such a two - site bond operator product gives zero if in the initial spin configuration all involved spin effective lattice sites refer to deconfined @xmath58 spinons .
such rules follow naturally from the definition of the subspace where the operators of eqs .
( [ g - s1+general])-([f - s1-operator ] ) act onto .
the main criterion is that such operators have been constructed to inherently generating a faithful representation provided that the corresponding state ( [ phi-+-s1 ] ) represents the @xmath1010 configuration state and hence is identical to that given in eq .
( [ phi-+ ] ) with the same value of @xmath1166 .
for instance , the second rule results from in all partitions of the summation on the right - hand side of eq .
( [ phi-+ ] ) a site of the spin effective lattice being linked to exactly only one site .
indeed , in a given partition no two - site bonds join the same site . the exact dependence on the two - site bond length @xmath1101 of the absolute value @xmath1149 of the coefficients appearing in eqs .
( [ phi-+ ] ) and ( [ g - s1+sum - rule ] ) remains an open problem .
however , if one assumes that @xmath1149 has the following simple power - law dependence on that length , @xmath1167 the two - site - bond - length expression ( [ xi - l ] ) and normalization condition ( [ g - s1+sum - rule ] ) alone imply @xmath1168 be given by , @xmath1169^{-\alpha_{s1 } } } \ , ; \hspace{0.50 cm } d=2 \ , , \nonumber \\ c^2 & = & { a_s^{2\alpha_{s1}}\over 2\sum_{n_1}(1 + 2n_d)^{-2\alpha_{s1 } } } \ , ; \hspace{0.50 cm } d=1 \ , .
\label{c - d-1 - 2}\end{aligned}\ ] ] here the summations run over the range of @xmath1073 and @xmath1074 ( and @xmath1106 ) values given in eq . ( [ link - sum ] ) for the model on the square ( and 1d ) lattice .
an expression of the general form ( [ c - l - d ] ) would imply that the dependence of @xmath1149 on @xmath10 and @xmath194 occurred only through that of the exponent @xmath1170 . of the primary two - site bonds of eq .
( [ rela - weight ] ) for the model on the square lattice as a function of the exponent @xmath1171 for the range @xmath1172 .
that relative weight refers to the simple power - law expressions given in eqs .
( [ c - l - d ] ) and ( [ c - d-1 - 2 ] ) for the coefficients @xmath1149 appearing in the summations of eq .
( [ g - s1+general ] ) .
the physical range of the exponent @xmath1171 corresponds typically to @xmath1173 so that the primary two - site bonds are behind most of the local @xmath64 fermion spectral weight .
for @xmath1174 the ratio @xmath1175 is larger than @xmath1176 and approaches quickly the unit upon further increasing @xmath1171.,width=264,height=132 ] for the coefficients @xmath1149 power - law expressions given in eqs .
( [ c - l - d ] ) and ( [ c - d-1 - 2 ] ) , the ratio of the spectral weight of a primary two - site bond over the weight of the corresponding two - site bond family , @xmath1177 } \vert h_{g}\vert^2 } = 2d\,\vert h_0\vert^2 \ , , \hspace{0.50 cm } z=1d,2d \ , , \label{ratios-2 - 1}\ ] ] is given by , @xmath1178^{-\alpha_{s1 } } } \ , ; \hspace{0.15 cm } d=2 \ , , \nonumber \\
r_{1d } & = & { 1\over\sum_{n_1}(1 + 2n_d)^{-2\alpha_{s1}}}\ , ; \hspace{0.15 cm } d=1 \ , .
\label{rela - weight}\end{aligned}\ ] ] the ratios of eq .
( [ rela - weight ] ) refer to the @xmath196 limit and are plotted in figs . 5 and 6 for the model on the square and 1d lattice , respectively , as a function of the exponent @xmath1171 for the range @xmath1172 .
for @xmath1171 slightly larger than @xmath1179 where @xmath1180 the ratios @xmath1181 ( @xmath36 ) and @xmath1175 ( @xmath37 ) are larger than @xmath1176 .
for @xmath196 the exact coefficients @xmath1149 are decreasing functions of the two - site bond length whose expressions most likely are not , at least for the whole two - site - bond - length range , of the simple power - law form given in eqs .
( [ c - l - d ] ) and ( [ c - d-1 - 2 ] )
. however the exact sum - rules ( [ g - s1+sum - rule ] ) together with the coefficients @xmath1149 being decreasing functions of the two - site bond length implies that at least for not a too small length the unknown exact coefficients @xmath1149 fall off as in eq .
( [ c - l - d ] ) and the corresponding ratios of eq .
( [ ratios-2 - 1 ] ) have a behavior similar to that shown in figs . 5 and 6 .
hence for the model on the square lattice one expects that @xmath1182 so that the ratio @xmath1175 is larger than @xmath1176 and the primary two - site bonds are behind most of the spectral weight of the local @xmath64 fermion , alike for the 1d lattice for @xmath1183 .
an important point is that at @xmath146 , provided that @xmath196 one may have for the model on the square lattice a ground state with both long - range antiferromagnetic order and @xmath64 fermion spinon pairing with @xmath1062-wave symmetry . indeed for values of the two - site bond length @xmath1101 of eq .
( [ xi - l ] ) not too small the absolute value @xmath1149 of the coefficients @xmath1150 of the local @xmath64 fermion operators of eq .
( [ f - s1-operator ] ) is expected to fall off as @xmath1184 .
provided that for the @xmath146 , @xmath346 , and @xmath12 absolute ground state the value of the exponent @xmath1171 is approximately in the range @xmath1185 , the corresponding spin - singlet spinon @xmath64 bond pairing of the @xmath1186 two - spinon @xmath64 bond particles associated with such operators can for @xmath10 not too small have @xmath1062-wave symmetry yet the corresponding spin occupancy configurations have long - range antiferromagnetic order @xcite .
note that at @xmath146 the spin effective lattice equals the original lattice .
the upper value @xmath1187 is that obtained from numerical results on spin - singlet two - spin bonds @xcite . in turn
, it is expected that for a well - defined domain of finite @xmath194 values the two - site bond lengths @xmath1101 of eq .
( [ xi - l ] ) and corresponding absolute values @xmath1149 of the coefficients @xmath1150 of the local @xmath64 fermion operators of eq .
( [ f - s1-operator ] ) are such that the corresponding @xmath64 fermion spinon pairing refers to a short - range spin order rather than to a long - range antiferromagnetic order as that occurring at @xmath146 @xcite .
the one- and two - electron subspace as defined in this paper is spanned by states whose numbers obey the approximate selection rules of eq .
( [ srs-0-ss ] ) .
such states have @xmath64 fermion occupancies with none , one , and two @xmath64 effective - lattice unoccupied sites .
there are several configuration states with one or two unoccupied sites in that lattice .
those refer to different positions of such sites associated with different local @xmath64 fermion occupancy configurations .
we recall that the corresponding @xmath64 fermion hole motion is independent of that of the rotated electrons that singly occupy sites of the original lattice relative to its remaining sites .
the latter are instead described by the motion of the @xmath52 fermions , whose occupancy configurations correspond to the state representations of the hidden global @xmath5 symmetry . in turn , the configuration states generated by the motion around in the @xmath64 effective lattice of the @xmath64 fermion holes associated with that lattice unoccupied sites refer to state representations of the global spin @xmath11 symmetry . the structure of the @xmath1010 configuration states is simpler than that of the @xmath1011 configuration states .
a key property however simplifies the study of the latter states : there is a one - to - one correspondence between the occupancies of the @xmath540 occupied sites of the @xmath64 effective lattice of the partitions of a @xmath1010 configuration state and those of the occupied sites of the @xmath64 effective lattice of partitions of the @xmath1011 configuration states with the same number @xmath540 of local @xmath64 fermions .
the point is that except for a change in the real - space coordinates of some of the sites , which accounts for the presence of one or two unoccupied sites , the two - site bond configurations of the sites of the spin effective lattice that are in one - to - one correspondence with each other remain unaltered .
indeed under suitable boundary conditions accounting for the presence of one or two unoccupied sites , the topology of the spinon occupancy configurations associated with the @xmath540 occupied sites of the @xmath64 effective lattice remains unaltered . of the primary two - site bonds of eq .
( [ rela - weight ] ) for the 1d model as a function of the exponent @xmath1171 for the range @xmath1172 .
the ratio @xmath1181 approaches the unit upon increasing the value of @xmath1171 more quickly than the ratio plotted in fig . 5 for the model on the square lattice .
for @xmath1189 integer the relative weight reads @xmath1190 $ ] where @xmath1191 is a riemann s zeta function .
for instance , @xmath1192 for @xmath1193 and @xmath1194 for @xmath1195.,width=264,height=132 ] for simplicity in this paper we omit a detailed study of the @xmath1011 configuration states .
the main result is that the slightly changed @xmath64 bond - particle operator expressions that account for the presence of one or two @xmath64 effective lattice unoccupied sites exactly obey to the same algebra as the corresponding @xmath64 bond - particle operators that act onto the @xmath1010 configuration state considered above . in appendix
d it is confirmed that for @xmath8 and the subspaces where the @xmath64 bond - particle operators of the one- and two - electron subspace are defined , upon acting onto the @xmath64 effective lattice such operators anticommute on the same site of that lattice , @xmath1196 and commute on different sites , @xmath1197 = [ g^{\dag}_{{\vec{r}}_{j},s1},g^{\dag}_{{\vec{r}}_{j'},s1 } ] = [ g_{{\vec{r}}_{j},s1},g_{{\vec{r}}_{j'},1 } ] = 0 \ , .
\label{g - non - local}\ ] ] here @xmath1136 .
that algebra implies that upon acting onto the @xmath64 effective lattice the @xmath64 bond - particle operators are hard - core like . an important consequence of such a property is that one can perform an extended jordan - wigner transformation @xmath1198 , whose phase operator @xmath1125 is given in eq .
( [ f - s1-operator ] ) .
it transforms the @xmath64 bond particles into @xmath64 fermions . as discussed in section v - d ,
that for the model on the square lattice each @xmath64 fermion has a flux tube of one flux quantum on average attached to it is consistent with the @xmath64 fermion and @xmath64 bond - particle wave functions obeying fermi and bose statistics , respectively .
the number of @xmath1199 configuration states is @xmath1200 = n_{s1}+1 $ ] .
those correspond to one - electron excited states .
the @xmath64 fermion occupancy configurations of ref .
@xcite concerning @xmath1199 excited states generated by application onto @xmath601 and @xmath12 ground states of one - electron operators can be expressed as suitable superpositions of such @xmath1200 = n_{s1}+1 $ ] configuration states . on the other hand , the @xmath64 fermion momentum occupancy of a @xmath1163 spin - triplet ( and spin - singlet ) excited state is described by a suitable superposition of a set of @xmath1201\,[n_{s1}+1]/2 $ ] configuration states of spin @xmath1202 with no @xmath436 fermions , so that @xmath979 ( and of vanishing spin @xmath841 and a single @xmath436 fermion so that @xmath977 . )
application of a local @xmath64 fermion creation operator of real - space coordinate @xmath160 onto such a ground state gives zero for most such configuration states . in turn ,
@xmath1163 configuration states whose two unoccupied sites refer to the configuration plotted in fig .
4 are transformed onto the @xmath1010 configuration state with one more @xmath64 fermion than the initial state .
this is so provided that the real - space coordinate of the corresponding central unoccupied site on the right - hand side of that figure coincides with that of the applied operator .
besides the unphysical processes excluded by the above three rules of @xmath1010 configuration states and similar corresponding rules of @xmath1011 configuration states , one finds that at spinon number @xmath49 fixed value application of a local @xmath64 fermion creation operator onto @xmath1203 configuration states gives always zero . in turn , application of a local @xmath64 fermion annihilation operator whose real - space coordinate is that of an occupied site of the @xmath64 effective lattice of a @xmath1188 configuration state transforms it into a state with two more unoccupied sites than the initial state .
this paper is a needed intermediate step between the extended global @xmath3 symmetry recently found in ref .
@xcite for the hubbard model on any bipartite lattice and the results on the square - lattice quantum liquid presented in ref .
the studies of that reference address such a symmetry physical consequences for the model on the square lattice described in terms of that quantum liquid of @xmath52 and @xmath64 fermions .
specifically , here we introduce a general operator description valid for both the hubbard model on the 1d and square lattice in terms of three types of elementary quantum objects whose occupancy configurations correspond to state representations of the model new - found extended global @xmath3 symmetry .
the description introduced here refers to the model full hilbert space .
its starting point is a unitary transformation that generates rotated electrons from the electrons .
it is such that rotated - electron double and single occupancy are good quantum numbers for @xmath8 .
the complete set of states associated with that description are generated by occupancy configurations of spinless and @xmath13-spinless @xmath52 fermions , spin-@xmath9 spinons , and @xmath13-spin-@xmath9 spinons .
those refer to three degrees of freedom of related rotated - electron occupancy configurations .
indeed the @xmath52 fermion and spinon occupancy configurations describe the hidden @xmath5 symmetry and spin @xmath11 symmetry , respectively , degrees of freedom of the rotated - electron singly occupied sites . in turn , the @xmath52 fermion hole and @xmath13-spinon occupancy configurations describe the hidden @xmath5 symmetry and @xmath13-spin @xmath11 symmetry , respectively , degrees of freedom of the rotated - electron doubly occupied and unoccupied sites . such hidden @xmath5 symmetry and two @xmath11 symmetries are those contained in the model extended global @xmath6/z_2 ^ 2 $ ] symmetry .
the @xmath52 , @xmath13-spin , and spin effective lattices and corresponding @xmath52 , @xmath60 , and @xmath61 momentum bands , respectively , are concepts valid in the @xmath196 limit .
such effective lattices have been constructed to inherently providing an approximate representation for the hidden @xmath5 symmetry , spin @xmath11 symmetry , and @xmath13-spin @xmath11 symmetry degrees of freedom of the rotated - electron occupancy configurations that generate a complete set of momentum eigenstates .
such occupancy configurations generate representations of the @xmath6/z_2 ^ 2 $ ] symmetry group .
the physical interest of the description introduced in this paper refers mostly to the model in the one- and two - electron subspace , as defined in section v. however , the definition of that subspace and the confirmation of the conservation in it of a set of quantities of interest for the model on the square lattice physics require the use of general properties of the full - hilbert - space description constructed in this paper .
this fully justifies the interest of our studies , which in spite of their extension and often technical character are a needed step for the introduction of the square - lattice quantum liquid further investigated in ref .
@xcite . that simpler quantum problem refers to the hubbard model on the square lattice in the one- and two - electron subspace as defined in this paper .
one of the main results of this paper is the introduction of such a quantum liquid .
two accomplishments are the following .
( i ) the use of suitable rotated electrons as defined in section ii allows the extension of spinon representations associated with singly occupied sites to all @xmath8 values rather than only for @xmath206 , as for the usual schemes in terms of electron singly occupancy .
( ii ) such rotated electrons generate a description consistent with the extended new found model global @xmath6/z_2 ^ 2 $ ] symmetry , what reveals that the corresponding rotated - electron occupancy configurations that generate the state representations of that symmetry refer to three degrees of freedom rather than to two .
the occurrence of only two fully symmetrical and independent degrees of freedom associated with two @xmath11 symmetries was suggested from the previously assumed model global @xmath7/z_2 $ ] symmetry @xcite . for @xmath4
the model contains indeed such a symmetry , yet it is not its full global symmetry .
the accomplishment ( ii ) has important consequences . for @xmath145 and
excitation energy @xmath1021 it reveals the occurrence of two degrees of freedom for the @xmath401 physics .
those are associated with the spinless @xmath52 fermion hidden @xmath5 symmetry and spin-@xmath9 spinon @xmath11 symmetry , respectively .
this provides a new scenario for studies of the effects of doping on the physics of the hubbard model on the square lattice . within our description , which emerges from the complete model global symmetry
, doping creates @xmath52 fermion holes . in turn , the corresponding @xmath145 and @xmath12 ground states are spin - singlet states whose @xmath64 fermion band remains full and thus all sites of the corresponding @xmath64 effective lattice remain occupied , alike for the @xmath146 and @xmath12 ground state . as investigated elsewhere , the effects of doping can then be described in terms of @xmath52 - @xmath64 fermion residual interactions . for small and intermediate values of @xmath10 the explicit form of the electron - rotated - electron unitary operator
@xmath89 associated with the rotated - electron operators as defined in this paper remains an open problem .
fortunately , however , for a one- or two - electron operator @xmath108 the terms of the general expression ( [ oor ] ) containing commutators involving the related operator @xmath128 generate nearly no spectral weight .
hence one can reach a quite faithful representation of such a one- or two - electron operator @xmath108 by replacing it by the corresponding leading - order operator term @xmath967 of expression ( [ oor ] ) and writing it in terms of the @xmath52 and @xmath64 fermion operators .
( @xmath967 is a one- or two - rotated - electron operator whose expression in terms of rotated - electron creation and annihilation operators is the same as that of @xmath108 in terms of electron creation and annihilation operators . ) the physical reason why for one- and two - electron operators the operator terms of the general expression ( [ oor ] ) containing commutators involving the operator @xmath128 generate very little one- and two - electron spectral weight is that they are of higher order in terms @xmath52 and @xmath64 fermion elementary processes .
indeed , the interactions of such objects are residual and the @xmath52 and @xmath64 fermion leading - order elementary processes are generated by the leading - order operator @xmath967 of expression ( [ oor ] ) .
this is confirmed by expressing it in terms of @xmath52 and @xmath64 fermion operators @xcite .
although our operator description is compatible with and in part inspired in the exact solution of the 1d model , it accounts for the basic differences between the physics of the hubbard model on the 1d and square lattice , respectively .
for instance , in 1d the occurrence of an infinite set of conservations laws associated with the model integrability implies that the @xmath52 - @xmath64 fermion residual interactions refer only to zero - momentum forward - scattering .
they merely give rise to phase shifts whose expressions may be extracted from the ba solution .
this allows the introduction of a pseudofermion dynamical theory , which provides finite - energy spectral and correlation function expressions involving phase shifts @xcite . hence in 1d such interactions
do not involve interchange of energy and momentum .
in contrast , they do for the hubbard model on the square lattice , yet they are much simpler than the corresponding electronic correlations . indeed the quantum problem is non - perturbative in terms of electron operators .
it follows that in contrast to a 3d isotropic fermi liquid @xcite , rewriting the square - lattice quantum - liquid theory in terms of the standard formalism of many - electron physics is in general an extremely complex problem .
fortunately , such a quantum liquid dramatically simplifies when expressed in terms of the @xmath52 fermion and @xmath64 fermion operators @xcite .
the problem is simplest at @xmath146 for spin excitations for which the @xmath52 band remains full and the effects of the @xmath52 - @xmath64 fermion interactions are frozen .
the preliminary investigations of ref .
@xcite on the physical consequences of the model on the square lattice new found global symmetry in actual materials in terms of the @xmath52 and @xmath64 fermion description refer to @xmath146 .
the results of such investigations confirm that the description introduced in this paper is useful for the further understanding of the role plaid by the electronic correlations in the spin spectrum of the parent compound la@xmath427cuo@xmath428 @xcite . indeed
, it is quantitatively described in that paper by the corresponding spin spectrum of the square - lattice quantum liquid at @xmath425 and @xmath426 mev .
elsewhere further investigations on more complex @xmath145 and @xmath12 2d problems for which the effects of doping are accounted for in terms of @xmath52 - @xmath64 fermion residual inelastic interactions will be fulfilled .
ou preliminary studies of such 2d problems seem to provide evidence that upon addition of a weak three - dimensional uniaxial anisotropy perturbation to the square - lattice quantum liquid introduced in this paper , a short - range spin order associated with the @xmath64 fermion spinon pairing , which occurs for a range of finite @xmath194 values @xcite , might coexist for low temperatures and a well - defined range of hole concentrations with a long - range superconducting order .
i thank discussions with m. a. n. arajo , h. q. lin , p. d. sacramento , and m. j. sampaio , suggestions on the figures by n. m. r. peres , the support and hospitality of the beijing computational science research center , and the support of the portuguese fct grant ptdc / fis/64926/2006 .
the studies of ref .
@xcite profit from the exact solution of the 1d model whose connection to the present description is discussed in this appendix .
note though that no explicit relations to the rotated - electron operators as those given in eqs .
( [ fc+])-([c - up - c - down ] ) are derived in that reference . moreover , no relation to the @xmath5 symmetry contained in the model @xmath3 global symmetry is established .
the eigenvalue of the generator ( [ or - ope ] ) of the @xmath5 symmetry is one half the number of rotated - electron singly occupied sites @xmath30 .
such an eigenvalue plays an important role in the present description .
it determines the values of the number @xmath51 of @xmath52 fermions , @xmath49 of spinons , @xmath1204 $ ] of @xmath52 fermion holes , and @xmath1205 $ ] of @xmath13-spinons .
( for simplicity , in this appendix we omit the index @xmath36 from the site numbers @xmath233 , @xmath219 , and @xmath453 where @xmath1206 and @xmath1207 . ) the studies of this paper establish that the spinons ( and @xmath13-spinons ) that are not invariant under the electron rotated - electron transformation are part of spin ( and @xmath13-spin ) neutral @xmath59-spinon ( and @xmath59-@xmath13-spinon ) composite @xmath61 ( and @xmath60 ) bond particles .
those have a binding ( and an anti - binding ) character .
fortunately , within the @xmath38 limit associated with our description the problem of the internal degrees of freedom of such composite @xmath63 bond particles separates from that of their positions in the corresponding @xmath63 effective lattice .
thus , the only general result needed for the goals of this appendix is that the composite @xmath63 bond - particle operators and its @xmath63 effective lattice have been constructed to inherently such operators anticommuting on the same @xmath63 effective - lattice site , @xmath1209 and commuting on different sites , @xmath1210 = [ g^{\dag}_{x_{j},\alpha\nu},g^{\dag}_{x_{j'},\alpha\nu } ] = [ g_{x_{j},\alpha\nu},g_{x_{j'},\alpha\nu } ] = 0 \ , ; \hspace{0.5 cm } j\neq j ' \ , .
\label{g - non - local - an}\ ] ] furthermore , such operators commute with the @xmath52 fermion operators and operators corresponding to different @xmath63 branches also commute with each other . in this appendix
we confirm that the general hard - core algebra of eqs .
( [ g - local - an ] ) and ( [ g - non - local - an ] ) combined with the universal number expressions given in section iv leads to discrete momentum values for the @xmath52 and @xmath63 fermions that coincide with the quantum numbers of the exact solution .
this holds for the whole lws hilbert subspace that such a solution refers to .
it follows from the algebra ( [ g - local - an])-([g - non - local - an ] ) that one can perform an extended jordan - wigner transformation .
it transforms the @xmath63 bond particles into @xmath63 fermions with operators @xmath1211 . alike in the general expressions provided in eq .
( [ f - an - operators ] ) , such operators are related to the corresponding bond - particle operators as , @xmath1212 here , @xmath1213 where @xmath1214 is the phase given in eq .
( [ f - an - operators ] ) .
however , for 1d the coordinates @xmath1215 and @xmath1216 appearing in that equation vanish .
hence @xmath160 reduces to the real - space coordinate @xmath1217 of the @xmath63 bond particle in its @xmath63 effective lattice .
therefore , for 1d the phase @xmath1214 can for all @xmath63 branches have only the values @xmath1218 and @xmath1219 .
indeed , the relative angle between two sites of the @xmath63 effective lattice in a 1d chain can only be one of the two values .
then the @xmath63 phase factor of eq .
( [ jw - phi - an ] ) is such that , @xmath1220 where @xmath1221 .
the @xmath52 fermion operators have the anticommuting relations given in eq .
( [ albegra - cf ] ) . for 1d those read , @xmath1222 moreover , the @xmath63 fermion operators that emerge from the jordan - wigner transformation associated with eqs .
( [ jw - f+-an ] ) and ( [ jw - phi - an ] ) have similar anticommuting relations given by , @xmath1223 in addition , the @xmath52 fermion operators commute with the @xmath63 fermion operators and @xmath63 and @xmath437 fermion operators such that @xmath1224 also commute . for 1d the @xmath52
fermion operators of eq .
( [ fc+ ] ) labeled by the discrete momentum values @xmath1225 read , @xmath1226 the @xmath63 fermion operators of eq .
( [ f - an - operators ] ) are in 1d denoted by @xmath1227 .
they are labeled by the discrete momentum values @xmath1225 such that @xmath1228 .
those are the conjugate variables of the @xmath63 effective lattice real - space coordinates @xmath1229 . for subspaces for which the ratio @xmath1230 involving the number @xmath705 of sites of the @xmath63 effective lattice given in eqs .
( [ n * ] ) and ( [ n - h - an ] ) is finite for @xmath1231 such operators are given by , @xmath1232 note that @xmath1233 . in 1d
the phase factor @xmath1234 does not have any effect when operating before @xmath1211 .
it follows that in 1d the expression of the hamiltonian does not involve the phase @xmath449 .
moreover , expression of the 1d ground - state normal - ordered hamiltonian in terms of the @xmath52 and @xmath63 fermion operators reveals that such objects have zero - momentum forward - scattering only .
this is consistent with the integrability of the model in 1d . in the present
@xmath1231 limit it is associated with the occurrence of an infinite number of conservations laws @xcite . for the 1d model
the occurrence of such conservations laws is behind the set of @xmath63 fermion numbers @xmath725 being good quantum numbers @xcite .
this is in contrast to the model on the square lattice , for which such numbers are not in general conserved , yet the number @xmath1235 of confined @xmath13-spinons ( @xmath246 ) and spinons ( @xmath247 ) is . the jordan - wigner transformations phases @xmath449 have direct effects on the boundary conditions .
those determine the discrete momentum values @xmath1225 of both the @xmath52 and @xmath63 fermion operators of eqs .
( [ fc - q - x-1d ] ) and ( [ fan - q - x-1d ] ) , respectively . in 1d
the periodic boundary conditions of the original electron problem are ensured provided that one accounts for the effects of the jordan - wigner transformation on the boundary conditions of the @xmath52 fermions and @xmath63 fermions upon moving one of such objects around the chain of length @xmath368 once . as discussed in section iv , for both the model on the 1d and square lattices the rotated - electron occupancies of the sites of the original lattice separate into two degrees of freedom only .
those of the @xmath30 sites of the original lattice singly occupied by rotated electrons separate into ( i ) @xmath30 sites of the @xmath52 effective lattice occupied by @xmath52 fermions and ( ii ) @xmath30 sites of the spin effective lattice occupied by spinons .
those of the @xmath1236 $ ] sites of the original lattice doubly occupied and unoccupied by rotated electrons separate into ( i ) @xmath1236 $ ] sites of the @xmath52 effective lattice unoccupied by @xmath52 fermions and ( ii ) @xmath1236 $ ] sites of the @xmath13-spin effective lattice occupied by @xmath13-spinons .
consider for instance the @xmath59 sites of the spin ( and @xmath13-spin ) effective lattice referring to the occupancy configuration of one local @xmath59-spinon composite @xmath61 fermion ( and @xmath59-@xmath13-spinon composite @xmath60 fermion ) .
those correspond to the spin ( and @xmath13-spin ) degrees of freedom of @xmath59 sites of the original lattice .
the corresponding degrees of freedom associated with the global @xmath5 symmetry found in ref .
@xcite are described by @xmath59 sites of the @xmath52 effective lattice occupied ( and unoccupied ) by @xmath52 fermions .
an important point is that the deconfined spinons and deconfined @xmath13-spinons are not part of the jordan - wigner transformations that map the @xmath63 bond particles onto @xmath63 fermions .
it follows that when one @xmath52 fermion moves around its effective lattice of length @xmath368 it feels the effects of the jordan - wigner transformations through the sites of the spin and @xmath13-spin lattices associated with those of the @xmath61 and @xmath60 effective lattices occupied by @xmath61 and @xmath60 fermions , respectively .
indeed , we recall that the sites of the @xmath52 effective lattice on the one hand and those of the spin and @xmath13-spin effective lattices on the other hand correspond to the different degrees of freedom of rotated - electron occupancies of the same sites of the original lattice .
the @xmath52 fermions do not emerge from a jordan - wigner transformation .
moreover , each @xmath63 fermion corresponds to a set of @xmath59 sites of the original lattice different from and independent of those of any other @xmath437 fermion .
it then follows that upon moving around its @xmath63 effective lattice of length @xmath368 , a @xmath63 fermion feels only the jordan - wigner - transformation phases of its own lattice .
those are associated with both the @xmath63 fermions and @xmath63 fermion holes .
hence its discrete momentum values obey the following periodic or anti - periodic boundary conditions , @xmath1237^{\dagger } e^{i(\phi_{j+1,\alpha\nu}-\phi_{j,\alpha\nu})}\right\ } = e^{i\pi [ n_{a_{\alpha\nu}}-1 ] } = -e^{i\pi n_{a_{\alpha\nu } } } \ , .
\label{pbcanp}\ ] ] here the phase factor reads @xmath523 and @xmath82 for @xmath1238 $ ] even and odd , respectively .
the term @xmath82 in @xmath1238 $ ] can be understood as referring to the site occupied by the @xmath63 fermion moving around its effective lattice and must be excluded . for the @xmath63 fermions
the unoccupied sites of their @xmath63 effective lattice exist in their own right .
indeed , note that according to eq .
( [ jw - f+-an ] ) both the creation and annihilation operators of such objects involve the jordan - wigner - transformation phase @xmath449 . as a result ,
such a phase affects both the @xmath63 fermions and @xmath63 fermion holes .
that justifies why the phase factor @xmath1239}$ ] of eq .
( [ pbcanp ] ) involves all the @xmath1240 $ ] sites of the @xmath63 effective lattice .
the only exception is that occupied by the moving @xmath63 fermion .
hence it involves both the @xmath1241 $ ] sites occupied by the remaining fermions of the same @xmath63 branch and the corresponding @xmath1242 @xmath63 fermion holes .
in contrast , the @xmath52 fermions are only affected by the sites occupied by @xmath63 fermions . indeed ,
only the sets of @xmath59 sites of the spin ( and @xmath13-spin ) effective lattice associated with each occupied site of the @xmath61 ( and @xmath60 ) @xmath1243 effective lattices and the sites of the spin ( and @xmath13-spin ) effective lattice occupied by deconfined spinons ( and deconfined @xmath13-spinons ) correspond to sites of the original lattice whose degrees of freedom associated with the @xmath52 fermion @xmath5 symmetry are described by the occupancy configurations of the @xmath52 effective lattice .
however , the deconfined spinons ( and deconfined @xmath13-spinons ) do not undergo any jordan - wigner transformation .
thus due to the jordan - wigner - transformation phase @xmath1244 of each of the @xmath462 @xmath63 fermions at sites @xmath1245 of their @xmath63 effective lattice the @xmath52 fermion discrete momentum values are determined by the following periodic or anti - periodic boundary condition , @xmath1246 again , the phase factor on the right - hand side of eq .
( [ pbccp ] ) reads @xmath523 and @xmath82 for @xmath556 even and odd , respectively .
the above results imply that the discrete momentum values @xmath1225 of both @xmath52 and @xmath63 fermions have the usual momentum spacing @xmath1247 and read , @xmath1248 however , following the boundary conditions ( [ pbcanp ] ) [ and ( [ pbccp ] ) ] the numbers @xmath1249 ( and @xmath1250 ) where @xmath1251 ( and @xmath1252 ) appearing in this equation are not always integers .
they are integers and half - odd integers for @xmath1238 $ ] ( and @xmath556 ) even and odd , respectively .
furthermore , as a result of the periodic or anti - periodic character of such boundary conditions these numbers obey the inequality @xmath1253/2 $ ] for both @xmath1238 $ ] odd and even ( and the inequality @xmath1254/2 $ ] for @xmath556 even and @xmath1255/2\leq i^{c}_j \leq n_a/2 $ ] for @xmath556 odd ) . for the one- and two - electron subspace as defined in section v one can separate the numbers @xmath1256 and @xmath1250 of eq .
( [ q - j - f - repr ] ) in two terms corresponding to an integer number and a small deviation as @xmath1257 $ ] and @xmath1258 $ ] , respectively .
the corresponding @xmath52 and @xmath64 fermion discrete momentum values then read , @xmath1259 here @xmath1260 ( and @xmath1261 ) is given either by @xmath1262 or @xmath1263 $ ] ( and @xmath1264 or @xmath1265 $ ] ) for all @xmath1266 ( and @xmath1267 ) discrete momentum values of the @xmath52 ( and @xmath64 ) band whose momentum occupancy describes a given state .
the @xmath64 effective lattice length @xmath1268 where @xmath1269\,a$ ] is the @xmath64 effective lattice spacing .
importantly , the @xmath52 fermion and @xmath63 fermion discrete momentum values obtained from our @xmath1208 operational description of the quantum problem correspond to the ba quantum numbers of the exact solution . indeed , the discrete momentum values of eq .
( [ q - j - f - repr ] ) can be expressed as , @xmath1270 here @xmath1271 , @xmath1272 , and @xmath1273 are the exact - solution integers or half integers quantum numbers involved in eqs .
( 2.12a)-(2.12c ) of ref . @xcite and defined in the unnumbered equations provided below these equations .
( in the notation of that reference , @xmath1274 and @xmath1275 in @xmath1276 and @xmath1277 )
. moreover , the numbers on the right - hand side of the two inequalities given just above eq .
( 2.13a ) of that reference correspond to @xmath1278 and @xmath1279 , respectively .
it follows that these inequalities read @xmath1280 and @xmath1281 .
this is fully consistent with the above inequality @xmath1253/2 $ ] where @xmath1206 and @xmath705 is given in eq .
( [ n * ] ) . a careful comparison of the notations and definitions used in ref .
@xcite and here confirms that there is also full consistency between the even or odd character of the integer numbers @xmath1282 $ ] , @xmath1283 $ ] , and @xmath556 considered here and those that determine the integer of half - integer character of the quantum numbers @xmath1272 , @xmath1273 , and @xmath1271 , respectively , in that reference .
we emphasize that for the @xmath63 fermions the discrete momentum values @xmath1225 of eq .
( [ q - j ] ) are the eigenvalues of the translation generator in the presence of the fictitious magnetic field of eq .
( [ a - j - s1 - 3d ] ) . for 1d
it reads @xmath1284 .
hence the corresponding exact - solution quantum numbers are the eigenvalues of such a translator operator in units of @xmath465 . for 1d
the numbers @xmath725 of @xmath63 fermions are conserved .
thus that the discrete momentum values @xmath1225 of the @xmath52 and @xmath63 fermions are good quantum numbers is consistent with the momentum operator commuting with the unitary operator @xmath767 as defined in this paper . that operator generates exact @xmath8 energy and momentum eigenstates @xmath1285 of general form given in eq .
( [ lws - full - el ] ) . indeed , in 1d @xmath1286 .
such states have @xmath10-dependent energy eigenvalues and @xmath10-independent momentum eigenvalues .
the use of the exact solution of the 1d problem confirms that the momentum eigenvalues have the general form given in eq .
( [ p-1 - 2-el - ss ] ) . for 1d
they may be written as , @xmath1287\ , n_{\eta\nu } ( q_{j } ) + \pi\,m_{\eta,-1/2 } \ , .
\label{p}\ ] ] here @xmath1288 is the total number of @xmath13-spin - projection @xmath57 @xmath13-spinons and the distributions @xmath1289 and @xmath1290 are the eigenvalues of the operators @xmath1291 and @xmath1292 , respectively .
those have values @xmath523 and @xmath524 for occupied and unoccupied momentum values , respectively .
one may obtain the expression ( [ p ] ) from analysis of the 1d problem for @xmath206 without the use of ba .
the starting point of such a procedure is that for @xmath529 the electrons that singly occupy sites do not feel the on - site repulsion .
consistently , expression ( [ p ] ) is that also provided by the exact solution after the rapidities are replaced by the quantum numbers @xmath1271 , @xmath1272 , and @xmath1273 related to the discrete momentum values @xmath1225 by eq .
( [ q - j ] ) . that the physical momentum ( [ p ] ) is for @xmath8 additive in the @xmath52 and @xmath63 fermion discrete momentum values results from the latter being good quantum numbers whose occupancy configurations generate the energy eigenstates .
it follows from the direct relation to the thermodynamic bethe ansatz equations of ref .
@xcite that the @xmath52 fermions obtained here from the rotated electrons through eq .
( [ fc+ ] ) and the @xmath63 fermions that emerge from the jordan - wigner transformations of eq .
( [ jw - f+-an ] ) are for the 1d model the @xmath52 pseudoparticles and @xmath63 pseudoparticles , respectively , associated in ref .
@xcite with the ba quantum numbers .
the momentum quantum numbers of eq .
( [ q - j ] ) are precisely those given in eq .
( a.1 ) of ref .
@xcite and the numbers @xmath705 of eq .
( [ n * ] ) equal the numbers @xmath1293 defined by its eqs .
( b.6 ) , ( b.7 ) , and ( b.11 ) with the index @xmath52 replaced by @xmath13 .
furthermore , the @xmath13-spinons and spinons considered here are for 1d the holons and spinons of that reference , respectively . also the deconfined @xmath13-spinons and deconfined spinons are for 1d the yang holons and hl spinons , respectively , of ref .
@xcite .
we just confirmed that for 1d the discrete momentum values of the @xmath52 fermion operators @xmath1294 , and those of the @xmath63 fermion operators @xmath1227 equal the quantum numbers of the exact ba solution .
such a result was obtained in the @xmath1208 limit that the description considered in this paper and in ref .
@xcite refers to . within our description
the @xmath52 fermions emerge from the electron - rotated - electron unitary transformation , as given in eq .
( [ fc+ ] ) . in turn , the @xmath63 fermions emerge from that transformation and an extended jordan - wigner transformation , as given in eqs .
( [ f - an - operators ] ) and ( [ jw - f+-an ] ) .
however , such a connection corresponds to the quantum numbers only .
the relation of the @xmath52 and @xmath63 fermion operators to the exact solution of the 1d model remains an open problem .
the relation of the building blocks of our description to the original electrons is uniquely defined yet corresponds to a complex problem .
such building blocks are the @xmath52 fermions , @xmath13-spinons , and spinons . for @xmath206
the rotated electrons become electrons and the @xmath52 fermion creation operator @xmath162 becomes the quasicharge annihilation operator @xmath185 of ref .
therefore , in that limit the @xmath52 fermions are the holes " of the quasicharge particles of that reference . in turn , the spinons and @xmath13-spinons are associated with the local spin and pseudospin operators , respectively , of the same reference . the transformation considered in ref .
@xcite does not introduce hilbert - space constraints .
it follows that suitable occupancy configurations of the objects associated with the local quasicharge , spin , and pseudospin operators considered in that reference exist that generate a complete set of states .
however , in 1d only in the limit @xmath206 suitable occupancy configurations of such basic objects generate exact energy eigenstates .
the point is that rotated electrons as defined in this paper are related to electrons by a unitary transformation . and
such a transformation is such that for @xmath8 rotated - electron occupancy configurations of the same form as those that generate energy eigenstates for @xmath206 in terms of electron operators do generate energy eigenstates for finite values of @xmath10 .
the @xmath52 fermion , @xmath13-spinon , and spinon operators are related to the rotated - electron operators as the quasicharge , spin , and pseudospin operators of ref .
@xcite are related to electron operators .
importantly , note that the validity of the @xmath52 fermion , spinon , and @xmath13-spinon operational description constructed in this paper and in ref .
@xcite for the hubbard model on a square and 1d lattices is for the 1d problem independent of its relation to the exact solution . for the lws subspace that such a solution refers to
the validity of our operational description follows from the transformations behind it not introducing hilbert - space constraints .
such transformations correspond to explicit operator expressions in terms of rotated - electron operators . for the @xmath52
fermion operators it is given in eq .
( [ fc+ ] ) . for the spinon and @xmath13-spinon operators
they are provided in eqs .
( [ sir - pir])-([rotated - quasi - spin ] ) . and
the rotated - electron operators are related to the original electron operators by the unitary transformation @xmath576 .
its unitary operator @xmath106 is within our description uniquely defined .
for @xmath8 it has been constructed to inherently generating a complete set of energy eigenstates of the general form @xmath1285 . here
@xmath1295 is a complete set of suitably chosen @xmath206 energy eigenstates .
for 1d such states are such that @xmath1296 where @xmath1297 are states of general form given in eq .
( [ lws - full - el ] ) . in order to clarify the relation of the @xmath52 and @xmath63 fermion operators to the exact solution of the 1d model rather than
the so called coordinate ba @xcite it is convenient to consider the solution of the problem by an algebraic operator formulation . within it
the hwss or lwss of the @xmath13-spin and spin algebras are built up in terms of linear combination of products of several types of creation fields acting onto the hole or electronic vacuum , respectively @xcite .
the 1d model energy eigenstates that are hwss or lwss of these algebras are often called _
bethe states_. here we briefly discuss how in 1d and for @xmath1208 the @xmath52 and @xmath63 fermion operators emerge from the creation fields of the algebraic formulation of the bethe states . that the general description introduced in this paper for the hubbard model on the square and 1d lattices is consistent with the exact solution of the 1d problem at the operator level as well , confirms its validity for 1d .
this is the only motivation and aim of this appendix .
however , since that refers to a side problem of that studied in this paper , in the following discussion we skip most technical details that are unnecessary for its general goals
. nevertheless provided that our analysis of the problem is complemented with the detailed information provided in refs .
@xcite , the resulting message clarifies the main issues under consideration .
the algebraic formulation of the bethe states refers to the transfer matrix of the classical coupled spin model , which is the `` covering '' 1d hubbard model @xcite .
indeed , within the inverse scattering method @xcite the central object to be diagonalized is the quantum transfer matrix rather than the underlying 1d hubbard model .
the transfer - matrix eigenvalues provide the spectrum of a set of @xmath1298 $ ] conserved charges .
the creation and annihilation fields are labeled by the ba rapidities @xmath1299 .
those may be generally complex and are not the ultimate quantum numbers of the model .
many quantities are functions of such rapidities . for instance , the weights @xmath1300 and @xmath1301 considered in the derivation of ref .
@xcite satisfy the free - fermion condition @xmath1302 .
( a possible and often used parametrization is @xmath1303 and @xmath1304 . )
the reparametrization @xmath1305\,e^{2h(\lambda)}- [ b(\lambda)/a(\lambda)]\,e^{-2h(\lambda)}-u/2 $ ] where the constraint @xmath1306 is defined by the relation @xmath1307=[u/2]\,a(\lambda)b(\lambda)$ ] , plays an important role in the derivation in the context of the quantum inverse scattering method of the non - trivial boltzmann weights of the isotropic six - vertex model given in eq .
( 33 ) of ref .
@xcite .
the diagonalization of the charge degrees of freedom involves a transfer matrix of the form provided in eq .
( 21 ) of that reference .
its off - diagonal entries are some of the above mentioned creation and annihilation fields .
the commutation relations of such important operators play a major role in the theory .
they are given in eqs .
( 25 ) , ( 40)-(42 ) , ( b.1)-(b.3 ) , ( b.7)-(b.11 ) , and ( b.19)-(b.22 ) of the same reference .
the solution of the spin degrees of freedom involves the diagonalization of the auxiliary transfer matrix associated with the monodromy matrix provided in eq .
( 95 ) of ref .
again , the off - diagonal entries of that matrix play the role of creation and annihilation operators .
their commutation relations are given in eq .
( 98 ) of that reference .
the latter commutation relations correspond to the usual faddeev - zamolodchikov algebra associated with the traditional abcd form of the elements of the monodromy matrix . in turn
, the above relations associated with the charge monodromy matrix refer to a different algebra .
the corresponding form of that matrix was called abcdf by the authors of ref .
@xcite . as discussed in ref .
@xcite , the main reason why the solution of the problem by the algebraic inverse scattering method @xcite was achieved only thirty years after that of the coordinate ba @xcite is that it was expected that the charge and spin monodromy matrices had the same traditional abcd form @xcite .
this would be consistent with the occurrence of a spin @xmath11 symmetry and a charge ( and @xmath13-spin ) @xmath11 symmetry known long ago @xcite , associated with a global @xmath7/z_2 $ ] symmetry @xcite .
fortunately , the studies of ref .
@xcite used an appropriate representation of the charge and spin monodromy matrices .
its structure is able to distinguish creation and annihilation fields as well as possible _
hidden symmetries _ , as discussed by the authors of that reference .
a hidden symmetry beyond @xmath17 was indeed identified recently .
it is the global @xmath5 symmetry found in ref .
the studies of that reference reveal that for @xmath8 the model charge and spin degrees of freedom are associated with @xmath45 and @xmath11 symmetries , rather than with two @xmath11 symmetries , respectively .
the occurrence of such charge @xmath45 symmetry and spin @xmath11 symmetry is fully consistent with the different abcdf and abcd forms of the charge and spin monodromy matrices of eqs .
( 21 ) and ( 95 ) of ref .
@xcite , respectively . indeed , the former matrix is larger than the latter .
it involves more fields than expected from the global @xmath7/z_2 $ ] symmetry alone .
this is consistent with the global @xmath1308/z_2 $ ] symmetry of the model on the 1d and any other bipartite lattice @xcite .
our general description of the model on a square and 1d lattices takes such an extended global symmetry into account .
furthermore , for 1d it was also implicitly taken into account by the appropriate representation of the charge and spin monodromy matrices used in ref . @xcite .
such a consistency is a necessary condition for the @xmath52 and @xmath63 fermion operators emerging from the fields of the monodromy matrices upon diagonalization of the 1d problem for @xmath1208 .
initially the expressions obtained by the algebraic inverse scattering method for the bethe states include both wanted terms and several types of unwanted terms .
the latter terms are eliminated by imposing suitable restrictions to the rapidities .
such constrains lead to the ba equations .
ultimately they lead to the real integer and half - integer quantum numbers . in units of @xmath465
those are the discrete momentum values that label the @xmath52 and @xmath63 fermion operators . after solving the ba for both the charge and spin degrees of freedom ,
one reaches the charge rapidities @xmath1309 with @xmath1310 $ ] values such that @xmath1311 $ ] .
in addition , one reaches the spin rapidities @xmath1312 with @xmath1313 $ ] values such that @xmath1314 $ ] . here
@xmath1315 and @xmath1316 for a hws and @xmath1317 and @xmath1318 for a lws of both the @xmath13-spin and spin algebras .
however , the rapidities can be extended to non - bethe tower states provided that the number of their values are expressed in terms of @xmath26 and @xmath211 rather than of @xmath199 and @xmath1319 , respectively .
equations of the same form as those obtained by the coordinate bethe anstaz are reached by the algebraic operator formulation .
this is so provided that one introduces the charge momentum rapidities @xmath1320 and spin rapidities @xmath1321 given by @xcite @xmath1322\,e^{2h(\lambda_j)}= e^{ik_j
a}$ ] where @xmath1311 $ ] and @xmath1323 where @xmath1314/2 $ ] . before discussing the structure of the rapidities for @xmath1208
, we confirm that the numbers @xmath1310 $ ] and @xmath1313 $ ] of discrete charge momentum and spin rapidity values , respectively , provided by the ba solution are closely related to the occupancy configurations of the rotated electrons that are not invariant under the electron - rotated - electron unitary transformation . indeed ,
such numbers can be rewritten as , @xmath1324 = [ 2s_c + m^{co}_{\eta } ] \ , ; \hspace{0.50 cm } [ n_a/2-s_{\eta}-s_s ] = [ m^{co}_{\eta } + m^{co}_{s}]/2 \ , .
\label{nn - cc}\ ] ] here the numbers @xmath238 and @xmath239 are those of eq . ( [ m - l - sum ] ) .
@xmath30 is the number of elementary rotated - electron charges associated with the singly occupied sites .
moreover , @xmath1325 is the number of such charges associated with the rotated - electron doubly occupied sites whose occupancy configurations are not invariant under the above unitary transformation .
furthermore , @xmath1326/2=\sum_{\alpha\nu}\nu\,n_{\alpha\nu}$ ] is the number @xmath1327 of down spins ( and up spins ) of such rotated - electron doubly occupied sites plus the number @xmath1328 of down spins ( and up spins ) of the rotated - electron singly occupied sites whose occupancy configurations are not invariant under that transformation .
the structure of the charge momentum and spin rapidities simplifies in the @xmath1208 limit @xcite . then the @xmath1310 = [ 2s_c + m^{co}_{\eta}]$ ] charge momentum rapidity values separate into two classes . those correspond to @xmath30 and @xmath238 of these values , respectively .
the set of charge momentum rapidity values @xmath1320 such that @xmath1329 are real and are related to the integer or half - integer quantum numbers @xmath1330 of eq .
( [ q - j - f - repr ] ) .
those are such that the numbers @xmath1331\,i^c_j$ ] are the @xmath30 discrete momentum values occupied by the @xmath52 fermions , out of a total number @xmath191 of such values .
they corresponds to the @xmath30 elementary charges of the rotated electrons that singly occupy sites .
moreover , for @xmath1208 the @xmath1313 = [ m^{co}_{\eta } + m^{co}_{s}]/2 $ ] spin rapidity values separate into two classes .
those correspond to @xmath231 and @xmath232 of these values , respectively .
the point is that the @xmath231 down spins and @xmath231 up spins of the rotated electrons that doubly occupy sites and whose occupancy configurations are not invariant under the electron - rotated - electron unitary transformation combine with the @xmath238 elementary charges left over by the above separation of the @xmath1310 = [ 2s_c + m^{co}_{\eta}]$ ] charge momentum rapidity values .
therefore , @xmath231 spin rapidity values out of @xmath1326/2 $ ] combine with @xmath238 charge momentum rapidity values out of @xmath1332 $ ] .
this leads to @xmath231 new rapidity values associated with the @xmath13-spin singlet configurations .
those describe the @xmath13-spin degrees of freedom of the rotated electrons that doubly occupy sites and whose occupancy configurations are not invariant under the electron - rotated - electron unitary transformation . in turn
, the @xmath232 spin rapidity values left over by the separation of the @xmath1313 = [ m^{co}_{\eta } + m^{co}_{s}]/2 $ ] spin rapidity values describe the spin - singlet configurations associated with the rotated electrons that singly occupy sites and whose occupancy configurations are not invariant under the electron - rotated - electron unitary transformation .
the charge degrees of freedom of these rotated electrons are described by the above set of charge momentum rapidity values @xmath1320 such that @xmath1329 . according to the @xmath1208 results of ref .
@xcite , the @xmath231 @xmath13-spin rapidity values ( and @xmath232 spin rapidity values ) further separate into @xmath60 rapidities of length @xmath62 with @xmath412 values ( and @xmath61 rapidities of length @xmath62 with @xmath413 values ) .
obviously , the sum - rules @xmath1327 and @xmath1328 are obeyed .
the results of that reference reveal that @xmath61 rapidities of length @xmath671 are real and the imaginary part of the remaining branches of @xmath63 rapidities where @xmath88 has a simple form .
therefore , for @xmath1208 the original ba equations lead to a system of @xmath1333 $ ] ( infinite in the limit @xmath1231 ) coupled thermodynamic equations .
their solution gives the charge momentum rapidity values @xmath1320 such that @xmath1329 and the values @xmath1334 of the real part of the @xmath63 rapidities such that @xmath1335 .
those are given as a function of the occupancies of the real integer or half - integer quantum numbers @xmath1250 and @xmath1249 of eq .
( [ q - j - f - repr ] ) , respectively .
alike for the @xmath52 fermions , the quantum numbers @xmath1249 such that @xmath1331\,i^{\alpha\nu}_j$ ] are the @xmath462 discrete momentum values occupied by @xmath63 fermions , out of a total number @xmath705 of such values .
( note that here we used a notation suitable to the extended global symmetry of the problem and that in ref .
@xcite such quantum numbers are denoted as in eq .
( [ q - j ] ) with @xmath473 replaced by @xmath42 and the values @xmath1336 and @xmath1337 of the real part of the rapidities by @xmath1338 and @xmath1339 , respectively , where @xmath1340 plays the role of @xmath451 . ) summarizing the above discussion , in order to reach the @xmath52 and @xmath63 fermion operators the algebraic operator formulation of the diagonalization of the quantum problem starts by building up the bethe states in terms of linear combination of products of the above mentioned several types of creation fields acting onto a suitable vacuum .
the diagonalization of the charge and spin degrees of freedom involves the transfer matrices given in eqs .
( 21 ) and ( 95 ) of ref .
@xcite , respectively .
their off - diagonal entries are some of these creation and annihilation fields .
the different form of such matrices is consistent with the model global @xmath3 symmetry .
it is taken into account as well by our general description both for the model on the 1d and square lattices used in studies of the square - lattice quantum liquid of @xmath52 and @xmath64 fermions .
the creation and annihilation fields obey the very involved commutation relations given in eqs .
( 25 ) , ( 40)-(42 ) , ( 98 ) , ( b.1)-(b.3 ) , ( b.7)-(b.11 ) , and ( b.19)-(b.22 ) of ref .
@xcite and are labeled by rapidities .
those may be generally complex and are not the ultimate quantum numbers .
however , such creation and annihilation fields and their involved algebra generate expressions for the bethe states that include both wanted terms and several types of unwanted and unphysical terms .
indeed , they act onto an extended and partially unphysical hilbert space , larger than that of the model .
the unwanted and unphysical terms are eliminated by imposing suitable restrictions to the rapidities that change the nature of the fields .
for @xmath1208 they are replaced by the @xmath52 and @xmath63 fermion operators labeled by the real integer and half - integer quantum numbers of the diagonalized model .
hence the ba equations obtained by imposing suitable restrictions to the rapidities describe the relation between the rapidities and the ultimate quantum numbers associated with the @xmath52 and @xmath63 fermion operators .
in addition to emerging from the elimination of the unwanted and unphysical terms of the bethe states generated by the initial creation and annihilation fields , the @xmath52 and @xmath63 fermion operator algebra refers to well - defined subspaces .
those are spanned by energy eigenstates whose number of @xmath13-spinons , spinons , and @xmath52 fermions is fixed .
it is given by @xmath361 $ ] , @xmath1341 , and @xmath51 , respectively .
hence the number @xmath30 of rotated - electron singly occupied sites and the numbers @xmath192 and @xmath193 of sites of the @xmath13-spin and spin lattices , respectively , are fixed . as discussed in this paper , for both the model on 1d lattice
considered in this appendix and the model on the square lattice the @xmath711 vacuum of the lws subspace @xmath253 given in eq .
( [ vacuum ] ) is invariant under the electron - rotated - electron unitary transformation .
for @xmath36 the @xmath193 deconfined @xmath58 spinons of such a vacuum are the spins of @xmath193 spin - up electrons , the @xmath192 deconfined @xmath58 @xmath13-spinons refer to the @xmath192 sites unoccupied by electrons , and the @xmath1342 @xmath52 fermions describe the charge degrees of freedom of such electrons of the fully polarized state .
the corresponding lwss that span the above subspace refer to rotated electrons rather than to electrons .
they have the general form given in eq .
( [ lws - full - el ] ) . in it
the set of numbers @xmath725 obey the sum - rule associated with the expressions provided in eq .
( [ m - l - sum ] ) , the @xmath13-spin ( @xmath214 ) and spin ( @xmath215 ) can have values @xmath1343 such that @xmath1344 $ ] , and the set of numbers @xmath1345 of discrete momentum values of each @xmath63 band are well defined and given by eq .
( [ n * ] ) and ( [ n - h - an ] ) . as mentioned previously , each subspace with fixed values of @xmath28 and hence also with fixed values of @xmath361 $ ] and @xmath1341 , which the vacuum of eq .
( [ vacuum ] ) refers to , can be divided into smaller subspaces .
those have fixed values for @xmath28 , @xmath26 , and @xmath27 and hence also fixed values for @xmath238 and @xmath510 .
furthermore , the latter subspaces can be further divided into even smaller subspaces with fixed values for the set of numbers @xmath511 and @xmath512 , which obey the sum - rule of eq .
( [ m - l - sum ] ) .
in contrast to the initial creation and annihilation fields , the @xmath52 and @xmath63 fermion operators generate the bethe states free from unwanted and unphysical terms , are labeled by the quantum numbers of the diagonalized model , and their algebra does not refer to the full lws or hws subspace but instead to subspaces spanned by well - defined types of bethe states .
hence there is no contradiction whatsoever between the charge abcdf algebra @xcite and spin abcd traditional faddeev - zamolodchikov algebra @xcite associated with involved commutation relations of the initial fields and the anticommutation relations of the @xmath52 and @xmath63 fermion operators provided in eqs .
( [ albegra - cf-1d ] ) and ( [ 1d - anti - com-1d ] ) .
indeed , the initial creation and annihilation fields act onto an extended and partially unphysical hilbert space . in turn ,
the @xmath52 and @xmath63 fermion operators act onto well - defined subspaces of the model physical hilbert space . upon
acting onto such subspaces , the operators @xmath1294 of the @xmath52 fermions and @xmath1227 of the @xmath63 fermions have the expected simple anticommutation relations associated with those provided in eqs .
( [ albegra - cf-1d ] ) and ( [ 1d - anti - com-1d ] ) for the corresponding operators @xmath1346 and @xmath1211 , respectively .
moreover , @xmath52 fermion operators commute with the @xmath63 fermion operators and @xmath63 and @xmath437 fermion operators belonging to different @xmath1347 branches also commute with each other .
in this appendix we confirm that full information about the quantum problem can be achieved by defining it in the lws subspace spanned by the energy eigenstates that are both lwss of the @xmath13-spin and spin algebras . by
that we mean that all @xmath851-electron operator matrix elements between energy eigenstates such that at least one of them is a non - lws can be evaluated exactly in terms of a corresponding quantum problem involving another well - defined @xmath851-electron operator acting onto the lws subspace .
here @xmath1348 refers to one , two , or any other finite number of electrons .
hence the expression of the general @xmath851-electron operator under consideration reduces to an elementary creation or annihilation electronic operator ( @xmath872 ) or involves the product of two or more such elementary operators ( @xmath1349 ) .
in addition , we show here that the use of the model global symmetry provides a simple relation between the energy of any non - lws and the corresponding lws .
it follows that all contributions to the physical quantities from non - lwss can be evaluated by considering a related problem defined in the lws subspace .
the use of such a symmetry reveals that for hole concentration @xmath145 and spin density @xmath320 the ground state is always a lws of both the @xmath13-spin and spin algebras . for simplicity here we consider matrix elements between the ground state and a non - lws .
those appear in lehmann representations of zero - temperature spectral functions and correlation functions .
similar results can be obtained for matrix elements between any excited energy eigenstates .
expressions ( [ fc+])-([rotated - quasi - spin ] ) for the @xmath52 fermion , @xmath13-spinon , and spinon operators refer to the lws subspace . a similar representation can be used for instance for the hws subspace .
it is spanned by all energy eigenstates with @xmath1350 where @xmath747 .
( there are also two mixed subspaces such that @xmath1351 and @xmath1352 . )
the hws representation is suitable for canonical ensembles referring to electronic densities larger than one and negative spin densities . in that case
the ground states are hwss of both the @xmath13-spin and spin algebras .
we start by confirming that there is a well - defined @xmath851-electron operator @xmath1353 such that any matrix element @xmath1354 of a @xmath851-electron operator @xmath857 can be written as @xmath1355 . here
@xmath1356 is a non - lws of the @xmath13-spin algebra and/or spin algebra , @xmath1357 is the ground state of the hubbard model on the 1d or square lattice , and @xmath1358 is the lws that corresponds to the state @xmath1356 . within the two @xmath11 algebras
the latter state can be expressed as , @xmath1359 in this expression , @xmath1360 \ , , \label{calpha}\ ] ] is a normalization constant and @xmath1361 .
the @xmath13-spin flip ( @xmath214 ) and spin flip ( @xmath215 ) operators @xmath1362 are the off - diagonal generators of the corresponding @xmath11 algebras provided in eq .
( [ scs ] ) .
these operators remain invariant under the electron - rotated - electron unitary transformation .
thus as given in that equation they have the same expression in terms of electron and rotated - electron creation and annihilation operators .
as confirmed in section iv , for a hole concentration @xmath601 and spin density @xmath566 the ground state is a lws of both the @xmath13-spin and spin @xmath11 algebras .
thus it has the following property , @xmath1363 the operator @xmath1353 is then such that , @xmath1364 by the suitable use of eq .
( [ gs=0 ] ) , it is straightforward to show that the operator @xmath1353 is given by the following commutator , @xmath1365\,\bigl[\prod_{\alpha = \eta , s}({\hat{s}}_{\alpha})^{m^{de}_{\alpha,-1/2 } } , { \hat{o}}_{{\cal{n}}}\bigr ] \ , , \label{th}\ ] ] for @xmath1366 and/or @xmath1367 and by , @xmath1368 for @xmath1369 . if the commutator on the right - hand side of eq .
( [ th ] ) vanishes then the matrix element @xmath1354 under consideration also vanishes .
we denote by @xmath1370 the energy eigenvalue of the non - lws @xmath1356 and by @xmath1371 that of the corresponding lws @xmath1358 . to find
the relation between @xmath1370 and @xmath1371 one adds chemical - potential and magnetic - field operator terms to the hamiltonian ( [ h ] ) , what lowers its symmetry .
such operator terms commute with the hamiltonian ( [ h ] ) .
thus the rotated - electron occupancy configurations of all energy eigenstates correspond to state representations of its global symmetry for all densities .
moreover , the use of such commutation relations reveals that the energy eigenvalues @xmath1370 and @xmath1371 are related as , @xmath1372 here @xmath1373 and @xmath564 are the deconfined @xmath57 @xmath13-spinon energy of eq .
( [ energy - eta ] ) for @xmath145 and the deconfined @xmath57 spinon energy of eq .
( [ energy - s ] ) for @xmath320 , respectively .
it follows from eq .
( [ energies ] ) that @xmath1374 . for a hole concentration @xmath145 and spin density @xmath320
such an inequality can be shown to be consistent with the ground state being always a lws of both the @xmath13-spin and spin algebras .
on the other hand , the model global symmetry requires that @xmath1375 for the half - filing and zero - magnetization absolute ground state .
such a requirement is fulfilled : for half filling one has that @xmath1376 .
the value @xmath346 corresponds to the middle of the mott - hubbard gap . in turn
, the magnetic field @xmath569 vanishes for zero magnetization .
the absolute ground state corresponds to @xmath886 and @xmath1377 .
hence it is both a lws and a hws of the @xmath13-spin and spin algebras . within a lehmann representation
the @xmath851-electron spectral functions are expressed as a sum of terms , one for each excited energy eigenstate .
in spectral - function terms associated with non - lwss one can then replace the matrix element @xmath1354 by @xmath1378 .
this holds provided that the excited - state energy @xmath1370 is expressed as in eq .
( [ energies ] ) .
then the original @xmath851-electron spectral function can be written as a sum of spectral functions of well - defined @xmath851-electron operators of the form given in eqs .
( [ th ] ) or ( [ ol-0 ] ) , all acting onto the lws subspace .
these results confirm that full information about the present quantum problem can be obtained by defining it in the lws subspace .
in the following we provide further information about why the square - lattice quantum liquid corresponding to the hubbard model on the square lattice in the one- and two - electron subspace may be described only by @xmath52 and @xmath64 fermions on their @xmath52 and @xmath64 effective lattices , respectively .
some of our results apply to the 1d model as well . for simplicity , in this appendix
we limit our considerations to @xmath145 initial ground states and their excited states of energy below @xmath402 . however , similar results hold for the model in the one- and two - electron subspace spanned by the @xmath146 , @xmath346 , and @xmath12 ground state and its excited states whose number value ranges are those provided in eq .
( [ srs-0-ss ] ) .
in addition , below we address the issue concerning the states of eq .
( [ non - lws ] ) that belong to the one- and two - electron subspace being both energy and momentum eigenstates and provide their specific form in that subspace . according to the number value ranges of eqs .
( [ srs-0-ss ] ) and ( [ deltancs1 ] ) , the one- and two - electron subspace is for hole concentrations @xmath145 spanned by an initial @xmath12 ground state plus its excited states of energy @xmath1021 having either none @xmath979 or one @xmath977 spin - neutral four - spinon @xmath436 fermion .
@xmath977 spin - singlet excited states belonging to that subspace have no deconfined spinons .
one then finds from the use of eq .
( [ n - h - an ] ) for the @xmath1379 branch that @xmath1380 for the latter states , so that they have no holes in the @xmath436 momentum band and thus @xmath1381 .
this means that for such states the @xmath436 fermion occupies a @xmath436 band with a single vanishing momentum value .
since such a @xmath436 fermion meets the criterion of eq .
( [ invariant - v ] ) of invariance under the electron - rotated - electron unitary transformation , the only explicit effect of its creation is onto the numbers of occupied and unoccupied sites of the @xmath64 effective lattice and corresponding numbers of @xmath64 band @xmath64 fermions and @xmath64 fermion holes .
specifically , according to the expressions provided in eq .
( [ nas1-nhs1 ] ) , the deviations @xmath1382 and @xmath1383 generated by a state transition involving creation of one @xmath436 fermion lead to deviations in the number of @xmath64 fermions and @xmath64 fermion holes given by @xmath1384 and @xmath1385 , respectively . moreover , the ranges of eqs .
( [ srs-0-ss ] ) and ( [ deltancs1 ] ) confirm that such @xmath977 excited states have zero spin , @xmath841 .
according to eq .
( [ nas1-nhs1 ] ) , the number of holes in the @xmath64 band is @xmath1386 for such states , in contrast to @xmath1387 for the initial ground state . in turn , the number @xmath1388 of sites of the @xmath64 effective lattice remains unaltered .
following the annihilation of two @xmath64 fermions and creation of one @xmath436 fermion , two unoccupied sites appear in the @xmath64 effective lattice . as a result ,
two holes emerge in the @xmath64 band as well .
the emergence of these unoccupied sites and holes involves two virtual processes where ( i ) two @xmath64 fermions are annihilated and four deconfined spinons are created and ( ii ) the latter deconfined spinons are annihilated and the @xmath436 fermion is created .
hence the only explicit net effect of the creation of a single vanishing - energy and zero - momentum @xmath436 fermion is the annihilation of two @xmath64 fermions and corresponding emergence of two holes in the @xmath64 band and two unoccupied sites in the @xmath64 effective lattice .
therefore , in the case of the one- and two - electron subspace at excitation energy @xmath1021 one can ignore that object in the theory provided that the corresponding changes in the @xmath64 band and @xmath64 effective lattice occupancies are accounted for . within neutral @xmath64 fermion particle - hole processes of transitions between two excited states with a single @xmath436 fermion , two of the four spinons of such an object are used in the motion of @xmath64 fermions around in the @xmath64 effective lattice . indeed , such two spinons play the role of unoccupied sites of that lattice , consistently with the expression @xmath1389 obtained from eq .
( [ nas1-nhs1 ] ) for @xmath841 .
spin - singlet excitations of the model on the square lattice generated by application onto a @xmath567 and @xmath12 initial ground state of the operator @xmath1390 where @xmath501 and @xmath755 are the momenta of the two emerging @xmath64 fermion holes are neutral states that conserve @xmath28 , @xmath27 , and @xmath619 .
the implicit role of the @xmath436 fermion creation operator @xmath1391 is exactly canceling the contributions of the annihilation of the two @xmath64 fermions of momenta @xmath501 and @xmath755 to the commutator @xmath1392 $ ] of the square - lattice model @xmath64 translation generators in the presence of the fictitious magnetic field @xmath615 of eq .
( [ a - j - s1 - 3d ] ) .
this ensures that the overall excitation is neutral .
since the @xmath436 fermion has vanishing energy and momentum and the @xmath64 band and its number @xmath619 of discrete momentum values remain unaltered , one can effectively consider that the generator of such an excitation is @xmath1393 and omit the @xmath436 fermion creation operator .
its only role is ensuring that the overall excitation is neutral and the two components of the square - lattice model @xmath64 fermion microscopic momenta can be specified .
it follows that for the one- and two - electron subspace considered here the operators @xmath1393 , @xmath1394 , @xmath1395 , and @xmath1396 generate neutral excitations .
also the @xmath1397 deconfined spinons play the role of unoccupied sites of the @xmath64 effective lattice .
again , this is consistent with the expression @xmath1398 obtained from eq .
( [ nas1-nhs1 ] ) for @xmath979 . as given in eq .
( [ srs-0-ss ] ) , for @xmath145 and excitation energy @xmath1021 the one- and two - electron subspace @xmath903 allowed values are @xmath851-electron - operator dependent and given by @xmath1399 where @xmath918 for @xmath145 . for @xmath1400 one
has that @xmath1401 .
now in contrast to creation of a single @xmath436 fermion , a deviation @xmath1402 generated by a transition from the ground state to such @xmath1403 excited states may lead to deviations in the numbers of occupied and unoccupied sites of the @xmath64 effective lattice and corresponding @xmath64 fermion and @xmath64 fermion holes that do not obey the usual equality @xmath1404 .
indeed , in the present case @xmath1405 for @xmath1406 and @xmath1407 for @xmath1408 . hence according to the expressions provided in eq .
( [ nas1-nhs1 ] ) , such deviations lead to deviations in the numbers of occupied and unoccupied sites of the @xmath64 effective lattice and corresponding numbers of @xmath64 fermions and @xmath64 fermion holes . those read @xmath1409 $ ] and @xmath1410 , respectively .
it follows that the total number of sites and thus of discrete momentum values of the @xmath64 band may change under such transitions .
this leads to an additional deviation @xmath1411 $ ] . for one - electron excited states
one has that @xmath1406 and @xmath1405 . as a result , @xmath1412 and @xmath1413 . in turn , for @xmath979 two - electron excited states one has @xmath1408 and @xmath1407 .
thus @xmath1414 and @xmath1415 . for the @xmath64 fermion operators @xmath1416 and @xmath1417 , excitations that involve changes
@xmath1411 $ ] in the number of sites and discrete momentum values of the @xmath64 effective lattice and @xmath64 band , respectively , correspond to transitions between different quantum problems . indeed
, such operators act onto subspaces spanned by neutral states , which conserve @xmath28 , @xmath27 , and @xmath701 . in turn ,
the generator of a non - neutral excitation is in general the product of two operators .
the first operator adds sites to or removes sites from the @xmath64 effective lattice or adds discrete momentum values to or removes discrete momentum values from the corresponding @xmath64 momentum band .
such small changes account for the above deviations @xmath1411 $ ] .
the second operator is a @xmath64 fermion operator or a product of such operators appropriate to the excited - state subspace . also the vanishing momentum and energy @xmath1418 deconfined @xmath58 @xmath13-spinons ( or @xmath1419 deconfined @xmath57 @xmath13-spinons if one uses a hws representation suitable to @xmath318 ) are invariant under the electron - rotated - electron unitary transformation .
their creation or annihilation may be accounted for by small suitable changes in occupancies of the @xmath52 effective lattice and @xmath52 momentum band . for @xmath145 , excitation energy below @xmath402 , and the one- and two - electron subspace considered here
such deconfined @xmath58 @xmath13-spinons correspond to a single occupancy configuration associated with the @xmath13-spin vacuum @xmath255 of eq .
( [ vacuum ] ) . in turn
, the degrees of freedom of the rotated - electron occupancies of such @xmath1420 sites of the original lattice associated with the hidden @xmath5 symmetry refer to the unoccupied sites of the @xmath52 effective lattice of eq .
( [ nac - nhc ] ) and corresponding @xmath52 band holes .
hence the number @xmath1418 of deconfined @xmath58 @xmath13-spinons equals that of @xmath1421 unoccupied sites of the @xmath52 effective lattice and corresponding @xmath52 band holes .
this confirms that for @xmath145 and excitation energy @xmath1021 the deviations @xmath1422 originated from creation and annihilation of deconfined @xmath58 @xmath13-spinons have within the one- and two - electron subspace no effects on the physics other than the corresponding deviation @xmath1423 in the number of unoccupied sites of the @xmath52 effective lattice and @xmath52 band holes .
concerning the effects of the presence of a single @xmath424 fermion or of one or two deconfined @xmath57 or @xmath58 @xmath13-spinons in the excited states of the @xmath146 , @xmath346 , and @xmath12 ground state , similar results to those reported above concerning the effects of a single @xmath436 fermion or one or two deconfined @xmath57 or @xmath58 spinons apply . again , except for a fixed uniquely defined finite - energy gap for each type of excitation , which vanishes for spin excitations @xcite , the half - filling hubbard model on the square lattice in the one- and two - electron subspace as defined in section v can be described by a quantum liquid of @xmath52 and @xmath64 fermions . here
we provide the specific form that the momentum eigenstates of eq .
( [ non - lws ] ) have in the one- and two - electron subspace for @xmath145 and excitation energy @xmath1021 .
such states refer to a complete set of states in the full hilbert space .
corresponding simple expressions apply to the excited states of the @xmath146 , @xmath346 , and @xmath12 ground state that belong to the one- and two - electron subspace as defined in section v. in general the states of eq .
( [ non - lws ] ) are not energy eigenstates of the model on the square lattice .
fortunately , in the one- and two - electron subspace such momentum energy eigenstates are energy eigenstates .
this confirms the usefulness of both the general description introduced in this paper and the square - lattice quantum liquid that refers to the hubbard model in that subspace .
the @xmath64 band discrete momentum values @xmath1424 where @xmath1043 are the conjugate of the real - space coordinates @xmath869 of the @xmath64 effective lattice for which also @xmath1043 .
the same applies to the @xmath52 band discrete momentum values @xmath1424 and the @xmath52 effective lattice real - space coordinates @xmath869 where in both cases @xmath1425 .
( the latter lattice is identical to the original lattice . ) as discussed in section iv - b , the @xmath52 translation generators @xmath1426 commute with both the hamiltonian and momentum operator for the whole hilbert space .
this is why the @xmath52 band discrete momentum values are good quantum numbers . in turn , the @xmath64 translation generators @xmath557 in the presence of the fictitious magnetic field @xmath615 of eq .
( [ a - j - s1 - 3d ] ) do not commute in general with the hamiltonian of the hubbard model on the square lattice .
the one- and two - electron subspace as defined in section v is a subspace of type ( a ) considered in section iv - f .
hence , the @xmath48 local spin @xmath11 gauge symmetry implies that the @xmath64 band momenta are good quantum numbers for the hubbard model on the square lattice .
that is so provided that for the model in the one- and two - electron subspace the numbers @xmath539 and @xmath540 are conserved . since in that subspace those are indeed good quantum numbers , the unitarity of the operator @xmath89 then implies that in such a subspace the @xmath64 band momentum values are conserved for @xmath8 as well .
thus the hamiltonian of the hubbard model on the square lattice in the one- and two - electron subspace commutes with the @xmath64 translation generators @xmath1427 and @xmath1428 in the presence of the corresponding fictitious magnetic field @xmath615 . indeed in the neutral subspaces of the one- and two - electron subspace the @xmath64 translation generators
@xmath557 can be constructed to inherently commuting with both the hamiltonian and momentum operator of the hubbard model on the square lattice .
note however that this requires a suitable definition of the @xmath64 fermions in terms of all spin - effective lattice two - site bonds centered at the local @xmath64 fermion spatial coordinate that in the @xmath48 limit profits from the model local spin @xmath11 gauge symmetry . since in contrast to the @xmath52 fermions , the @xmath64 fermions have internal structure ,
how is the @xmath64 fermion momentum @xmath501 related to the two underlying spinons ?
deconfined spinons carry no momentum and are invariant under the electron - rotated - electron unitary transformation . on the other hand , within the lws representation of the spin @xmath11 algebra , for the states that span the one- and two - electron subspace for which @xmath539 may have the values @xmath958 , the spin - down spinon of the spin - singlet two - spinon @xmath64 fermion of momentum @xmath501 carries momentum @xmath501 and its spin - up spinon carries momentum @xmath1429 . in turn , within the hws representation of that algebra , its spin - down spinon carries momentum @xmath1429 and its spin - up spinon carries momentum @xmath501 . within the present lws representation , the spin - singlet two - spinon @xmath64 fermions of momenta ( i ) @xmath501 and ( ii )
@xmath1429 involve ( i ) a spin - down spinon of momentum @xmath501 and a spin - up spinon of momentum @xmath1429 and ( ii ) a spin - down spinon of momentum @xmath1429 and a spin - up spinon of momentum @xmath501 , respectively .
we emphasize that for @xmath950 states , one has that @xmath1430 in the momentum spectrum @xmath1431 , so that the total @xmath64 band momentum vanishes . as discussed elsewhere , the momenta @xmath501 or @xmath501 and @xmath755 of the @xmath64 band holes of @xmath1432 states can alternatively be associated with that of the annihilated @xmath64 fermion or corresponding spinons of the broken spinon pair .
the set of neutral energy eigenstates that for @xmath145 and excitation energy @xmath1021 span the subspaces of the one- and two - electron subspace are particular cases of the general momentum eigenstates @xmath107 of eq .
( [ non - lws ] ) .
the use of the general expression of such states given in that equation leads to the following corresponding general form for such energy eigenstates @xmath1433 , @xmath1434 = 1 , 2 , 4 \ , .
\label{non - lws - c - s1}\ ] ] the lws appearing in this equation refers to a particular form of the general lws of eq .
( [ lws - full - el ] ) and reads , @xmath1435 [ \prod_{{\vec{q}}\,'}f^{\dag}_{{\vec{q}}\,',s1}\vert 0_{s1};n_{a_{s}^d}\rangle ] [ \prod_{{\vec{q}}}f^{\dag}_{{\vec{q}},c}\vert gs_c;0\rangle ] \ , .
\label{lws - full - el - c - s1}\ ] ] here @xmath1436 , @xmath1437 , and @xmath1438 and @xmath533 are the creation operators of a @xmath48 @xmath64 fermion of momentum @xmath755 and @xmath52 fermion of momentum @xmath501 , respectively .
moreover , @xmath255 is the @xmath13-spin @xmath11 vacuum associated with @xmath188 deconfined @xmath58 @xmath13-spinons , @xmath256 is the spin @xmath11 vacuum associated with @xmath221 deconfined @xmath58 spinons , and @xmath259 is the @xmath52 fermion @xmath5 vacuum .
such three vacua are invariant under the electron - rotated - electron unitary transformation , refer to the model global @xmath20/z_2 ^ 2 = so(3)\times so(3)\times u(1)$ ] symmetry , and appear in the theory vacuum of eq .
( [ vacuum ] ) .
( in that equation , @xmath1439 . )
the goal of this appendix is to confirm the validity of the relations provided in eqs .
( [ g - local ] ) and ( [ g - non - local ] ) .
all the results given in the following apply both to the model on the square and 1d lattices .
the 1d expressions are readily obtained if one considers in the general expressions given below only @xmath1048 contributions and terms , together with the choice @xmath36 in the @xmath34-dependent quantities . here the bond index @xmath1062 is that appearing in the @xmath1440 summation of the operators @xmath1143 and @xmath1144 expression provided in eq .
( [ g - s1+general ] ) . in order to probe the relations of eqs .
( [ g - local ] ) and ( [ g - non - local ] ) , we consider without any loss of generality that the initial state is a @xmath1441 configuration state . indeed ,
application of a @xmath64 bond - particle creation operator onto @xmath1203 configuration states gives zero .
application of two @xmath64 bond - particle creation operators onto @xmath1163 configuration states gives zero as well .
fortunately , the discussions about @xmath1010 configuration states of section vi can be generalized to @xmath1441 configuration states .
one finds that application onto some of such configuration states of two two - site bond operators may not give zero .
furthermore , it turns out that for such @xmath1441 configuration states the presence of the four unoccupied sites in the @xmath64 effective lattice does not affect the spinon occupancy configurations of the @xmath540 @xmath64 bond particles so that their operator expressions are the same as for the @xmath1010 configuration state studied in section vi .
we start by combining the expressions given in eq .
( [ sir - pir ] ) with the algebraic relations provided in eqs .
( [ albegra - s - h])-([albegra - q - com ] ) , what readily leads to the following usual algebra for the spinon operators defined in eq .
( [ sir - pir ] ) , @xmath1442 @xmath1443 = [ s^{\pm}_{\vec{r}_j},s^{\pm}_{\vec{r}_{j'}}]=0 \ , , \label{albegra - s - com}\ ] ] for @xmath1136 and , @xmath1444 = \pm\delta_{j , j'}s^{\pm}_{\vec{r}_{j } } \ , .
\label{albegra - s - sz - com}\ ] ] it follows that the spinon operators @xmath1445 anticommute on the same site and commute on different sites . consistently with the rotated - electron singly - occupied site projector @xmath205 appearing in the expression of the spinon operators @xmath1445 and @xmath1446 provided in eq .
( [ sir - pir ] ) and the discussion of section iii on the justification of the validity of the concept of a spin effective lattice in the @xmath196 limit , the real - space coordinates @xmath160 of such operators can in that limit be identified with those of the latter lattice .
the corresponding operator index values @xmath1447 are thus those of that effective lattice . for the @xmath1441 configuration states under consideration the operators @xmath1037 ( and @xmath1139 ) that create ( and annihilate ) a @xmath64 bond particle at a site of the spin effective lattice of real - space coordinate @xmath1039 have the general form given in eq .
( [ g - s1+general ] ) both for the hubbard model on the 1d and square lattices .
the absolute value @xmath1149 of the coefficients @xmath1150 appearing in that equation decreases for increasing two - site bond length @xmath1101 and obeys the normalization sum - rule ( [ g - s1+sum - rule ] ) .
hence the expression of the @xmath64 bond - particle operator @xmath1448 involves the operators @xmath1143 and @xmath1144 of eq .
( [ g - s1+general ] ) , which create and annihilate , respectively , a superposition of @xmath1044 two - site bonds of the same type .
the expression of such two - site bond operators @xmath1145 and @xmath1146 is given in eq .
( [ g - s - l ] ) . in order to confirm the validity of eqs .
( [ g - local ] ) and ( [ g - non - local ] ) , we use eqs .
( [ g - s1+general])-([g - s - l ] ) to rewrite the anti - commutation relations of eq .
( [ g - local ] ) in terms of anti - commutators of two - site bond operators as follows , @xmath1449 @xmath1450 @xmath1451 for simplicity here we have used the abbreviated summation notation , @xmath1452 moreover , on using of the same equations , the commutation relations of eq .
( [ g - non - local ] ) can be expressed in terms of the commutators of two - site bond operators .
that leads to , @xmath1197 = \sum_{d , l , g}\sum_{d',l',g ' } h^*_g\,h_{g ' } [ b_{{\vec{r}}_j + \vec{r}_{d , l}^{\,0},s1,d , l , g}^{\dag } , b_{{\vec{r}}_{j ' } + \vec{r}_{d',l'}^{\,0},s1,d',l',g ' } ] \ , , \label{g - non - local - ap+-}\ ] ] @xmath1453 = \sum_{d , l , g}\sum_{d',l',g ' } h^*_g\,h^*_{g ' } [ b_{{\vec{r}}_j + \vec{r}_{d , l}^{\,0},s1,d , l , g}^{\dag } , b_{{\vec{r}}_{j ' } + \vec{r}_{d',l'}^{\,0},s1,d',l',g'}^{\dag } ] \ , , \label{g - non - local - ap++}\ ] ] @xmath1454 = \sum_{d , l , g}\sum_{d',l',g ' } h_g\,h_{g ' } [ b_{{\vec{r}}_j + \vec{r}_{d , l}^{\,0},s1,d , l , g } , b_{{\vec{r}}_{j ' } + \vec{r}_{d',l'}^{\,0},s1,d',l',g ' } ] \ , , \label{g - non - local - ap--}\ ] ] where @xmath1136 .
according to the studies of section vi - c , three rules follow from the definition of the subspace where the operators of eqs .
( [ g - s1+general])-([g - s - l ] ) act onto .
the evaluation of the anti - commutators and commutators of the two - site bond operators on the right - hand side of eqs .
( [ g - local - ap+-])-([g - local - ap ] ) and ( [ g - non - local - ap+-])-([g - non - local - ap ] ) , respectively , relies on both such rules and the algebra given in eqs .
( [ albegra - s - p - m])-([albegra - s - sz - com ] ) of the spinon operators @xmath1445 and @xmath1446 of eq .
( [ sir - pir ] ) , which are the building blocks of the two - site bond operators of eq .
( [ g - s - l ] ) . fortunately , according to eqs .
( [ albegra - s - p - m])-([albegra - s - sz - com ] ) the spinon operators @xmath1445 obey the usual algebra : they anticommute on the same site of the spin effective lattice and commute on different sites . the two - site bond operators of eq .
( [ g - s - l ] ) can be rewritten as , @xmath1455 s^-_{\vec{r}_2 } -
\left[{1\over 2}+s^{x_3}_{\vec{r}_2}\right ] s^-_{\vec{r}_1}\right ) \ , , \label{g - s - l - simple}\ ] ] and @xmath1456 where recalling that the real - space coordinate of their two - site bond center reads @xmath1457 the real - space coordinates @xmath1458 and @xmath1459 are given by , @xmath1460 the evaluation of the anti - commutators and commutators of the two - site bond operators on the right - hand side of eqs . ( [ g - local - ap+-])-([g - local - ap ] ) and ( [ g - non - local - ap+-])-([g - non - local - ap ] ) , respectively , then relies on straightforward manipulations based on eqs .
( [ albegra - s - p - m])-([albegra - s - sz - com ] ) and the use of the three rules given in section vi - c .
the latter exclude unphysical processes .
this is equivalent defining the subspace that the operator algebra under consideration refers to .
for the two general anti - commutators needed to evaluate the two - site bond operators on the right - hand side of eq .
( [ g - local - ap+- ] ) we find the following expressions , @xmath1461\ } \ , , \label{anti - com - b+b - on}\end{aligned}\ ] ] @xmath1462 where @xmath1463 , @xmath1464 , @xmath1465 , and @xmath1466 . indeed , according to the second rule of section vi - c only general operators @xmath1467 whose two - site bond operators @xmath1468 and @xmath1469 do not join sites or join both sites of the spin effective lattice lead to wanted and physical spin configurations .
moreover , in the initial configuration onto which the four - site operator pairs of two - site bond operators appearing in eq .
( [ anti - com - b+b - on ] ) act one has according to the first and second rules that the two sites of real - space coordinates @xmath1458 and @xmath1459 either ( i ) refer to two deconfined @xmath58 spinons or ( ii ) are linked by a bond . in turn , the anti - commutator of eq .
( [ anti - com - b+b - off ] ) has the same form as those on the right - hand side of eq .
( [ g - local - ap+- ] ) .
hence one has that @xmath1470 yet @xmath1471 in eq .
( [ variables ] ) so that the restrictions imposed by the third rule are fulfilled . since we find that @xmath1472 and each of such operators is given by one half the operator on the right - hand side of eq .
( [ anti - com - b+b - off ] ) , when the two sites of real - space coordinates @xmath1458 and @xmath1459 ( and @xmath1473 and @xmath1474 ) are linked by a bond ( and refer to two deconfined @xmath58 spinons ) one must consider both initial configurations for which the sites of real - space coordinates @xmath1473 and @xmath1474 ( and @xmath1458 and @xmath1459 ) ( i ) are linked by a bond and ( ii ) refer to two deconfined @xmath58 spinons . in turn , when the two sites of real - space coordinates @xmath1458 and @xmath1459 ( and @xmath1473 and @xmath1474 ) refer to two deconfined @xmath58 spinons ( and are linked by a bond ) one must consider only initial configurations for which the sites of real - space coordinates @xmath1473 and @xmath1474 ( and @xmath1458 and @xmath1459 ) are linked by a bond ( and refer to two deconfined @xmath58 spinons ) .
it then follows from analysis of the operator expression on the right - hand side of eq .
( [ anti - com - b+b - on ] ) that when in the initial configuration the two sites @xmath1458 and @xmath1459 refer to two deconfined @xmath58 spinons the operator term @xmath1475(1/2+s^{x_3}_{\vec{r}_i})^2 $ ] transforms that configuration onto itself whereas the remaining operator terms give zero . in turn , when in the initial configuration the two sites @xmath1458 and @xmath1459 are linked and correspond to a bond configuration , the operator terms @xmath1475\{(1/2+s^{x_3}_{\vec{r}_i})^2 -s^+_{\vec{r}_i}s^-_{\vec{r}_{\bar{i}}}(1/2+s^{x_3}_{\vec{r}_{\bar{i}}})\}$ ] transform that configuration onto itself whereas the remaining operator terms give zero . on the other hand , when acting onto the above initial configurations the operator on the right - hand side of eq .
( [ anti - com - b+b - off ] ) gives always zero so that when acting onto the subspace that the operators of eqs .
( [ g - s1+general])-([g - s - l ] ) refer to the anti - commutators provided in eqs .
( [ anti - com - b+b - on ] ) and ( [ anti - com - b+b - off ] ) simplify and are given by , @xmath1476 where @xmath1465 and @xmath1466 . next concerning the two general anti - commutators needed to evaluate the two - site bond operators on the right - hand side of eq .
( [ g - local - ap++ ] ) we find the following expressions , @xmath1477 @xmath1478 where as above @xmath1463 , @xmath1464 , @xmath1465 , and @xmath1466 .
it follows from analysis of the operator on the right - hand side of eq .
( [ anti - com - b+b+on ] ) that , according to the first and second rules reported in section vi - c , when in the initial spin configuration both the sites of real - space coordinates @xmath1458 and @xmath1459 ( i ) refer to two deconfined @xmath58 spinons and ( ii ) are linked by a bond , application of that operator onto such a configuration gives zero . in turn the anti - commutator of eq .
( [ anti - com - b+b+off ] ) is of the form of those on the right - hand side of eq .
( [ g - local - ap++ ] ) so that in eq .
( [ variables ] ) one has that @xmath1470 yet @xmath1471 and thus the third rule of section vi - c applies . since we find that @xmath1479 and each of such operators is given by one half the operator on the right - hand side of eq .
( [ anti - com - b+b+off ] ) , concerning the latter operator when the sites of real - space coordinates @xmath1458 and @xmath1459 ( and @xmath1473 and @xmath1474 ) are linked by a bond one must consider both initial configurations for which the sites of real - space coordinates @xmath1473 and @xmath1474 ( and @xmath1458 and @xmath1459 ) ( i ) are linked by a bond and ( ii ) refer to two deconfined @xmath58 spinons . in turn , when the two sites of real - space coordinates @xmath1458 and @xmath1459 ( and @xmath1473 and @xmath1474 ) refer to two deconfined @xmath58 spinons one must consider only initial configurations for which the sites of real - space coordinates @xmath1473 and @xmath1474 ( and @xmath1458 and @xmath1459 ) are linked by a bond @xcite .
it follows then from analysis of the operator on the right - hand side of eq .
( [ anti - com - b+b+off ] ) that application of it onto any of such spin configurations gives zero .
a similar analysis for the two general anti - commutators needed to evaluate the two - site bond operators on the right - hand side of eq .
( [ g - local - ap ] ) leads to , @xmath1480 @xmath1481 where as above @xmath1463 , @xmath1464 , @xmath1465 , and @xmath1466 .
again analysis of the operator on the right - hand side of eq .
( [ anti - com - b - b - on ] ) reveals that according to the first and second rules of section vi - c when in the initial spin configuration both the sites of real - space coordinates @xmath1458 and @xmath1459 ( i ) refer to two deconfined @xmath58 spinons and ( ii ) are linked by a bond , application of that operator onto such a configuration gives zero . on the other hand , the anti - commutator of eq .
( [ anti - com - b - b - off ] ) is of the form of those on the right - hand side of eq .
( [ g - local - ap++ ] ) so that in eq .
( [ variables ] ) one has that @xmath1470 yet @xmath1471 and then the third rule applies . since we find that @xmath1482 and each of such operators is given by one half the operator on the right - hand side of eq .
( [ anti - com - b+b+off ] ) , concerning the latter operator when the sites of real - space coordinates @xmath1458 and @xmath1459 ( and @xmath1473 and @xmath1474 ) refer to two deconfined @xmath58 spinons one must consider both initial configurations for which the sites of real - space coordinates @xmath1473 and @xmath1474 ( and @xmath1458 and @xmath1459 ) ( i ) are linked by a bond and ( ii ) refer to two deconfined @xmath58 spinons . in turn , when the two sites of real - space coordinates @xmath1458 and @xmath1459 ( and @xmath1473 and @xmath1474 ) are linked by a bond one must consider only initial configurations for which the sites of real - space coordinates @xmath1473 and @xmath1474 ( and @xmath1458 and @xmath1459 ) refer to two deconfined @xmath58 spinons @xcite .
analysis of the operator on the right - hand side of eq .
( [ anti - com - b+b+off ] ) then reveals that application of it onto any of such spin configurations gives zero .
it then follows from the above results that when acting onto the subspace that the operators of eqs .
( [ g - s1+general])-([g - s - l ] ) refer to the anti - commutators provided in eqs .
( [ anti - com - b+b+on ] ) and ( [ anti - com - b+b+off ] ) and eqs .
( [ anti - com - b - b - on ] ) and ( [ anti - com - b - b - off ] ) simplify and read , @xmath1483 where @xmath1465 and @xmath1466 .
the use in eq .
( [ g - local - ap+- ] ) of the anti - commutators of eq .
( [ anti - com - b+b - on - off ] ) with the two - site bond operators related to those of eq .
( [ g - s - l ] ) by the expressions provided in eqs .
( [ g - s - l - simple ] ) and ( [ variables ] ) leads to , @xmath1484 which is the first relation of eq .
( [ g - local ] ) . to perform the summation of eq .
( [ g - local - ap+-final ] ) the sum - rule ( [ g - s1+sum - rule ] ) is used .
furthermore , the use of the anti - commutators of eq .
( [ anti - com - b+-b+-on - off ] ) in eqs .
( [ g - local - ap++ ] ) and ( [ g - local - ap ] ) leads to the remaining relations of eq .
( [ g - local ] ) .
the evaluation of the commutators of eq .
( [ g - non - local ] ) by the use of the expressions given in eqs .
( [ g - non - local - ap+-])-([g - non - local - ap ] ) is much simpler .
first it is simplified by the property that two - site bonds belonging to @xmath64 bond - particle operators with different real - space coordinates are always different .
second the evaluation of such commutators further relies on straightforward manipulations based on eqs .
( [ albegra - s - p - m])-([albegra - s - sz - com ] ) , which lead directly to , @xmath1485 = [ b_{\vec{r}_1,\vec{r}_2}^{\dag},b_{\vec{r}_{1'},\vec{r}_{2'}}^{\dag } ] = [ b_{\vec{r}_1,\vec{r}_2},b_{\vec{r}_{1'},\vec{r}_{2 ' } } ] = 0 \ , , \label{com - all}\ ] ] for @xmath1465 and @xmath1466 . finally , the use in eqs .
( [ g - non - local - ap+-])-([g - non - local - ap ] ) of the commutators of eq .
( [ com - all ] ) with the two - site bond operators related to those of eq .
( [ g - s - l ] ) by the expressions provided in eqs .
( [ g - s - l - simple ] ) and ( [ variables ] ) leads readily to the commutation relations provided in eq .
( [ g - non - local ] ) .
e. h. lieb , f. y. wu , phys .
( 1968 ) 1445 . m. j. martins , p. b. ramos , nucl .
b 522 ( 1998 ) 413 .
minoru takahashi , progr .
phys 47 ( 1972 ) 69 .
e. h. lieb , phys .
62 ( 1989 ) 1201 . j. m. p. carmelo , nucl
b 824 ( 2010 ) 452 ; 840 ( 2010 ) 553 , erratum . z. y. meng , t. c. lang , s. wessel , f. f. assad , a. muramatsu , nature 464 ( 2010 ) 847 .
r. jrdens , n. strohmaier , k. gnter , h. moritz , t. esslinger , nature 455 ( 2008 ) 204 .
d. jaksch , p. zoller , ann .
( 2005 ) 52 .
o. j. heilmann , e. h. lieb , ann . n. y. acad .
172 ( 1971 ) 584 . c. n. yang , s. c. zhang , mod .
b 4 ( 1990 ) 759 ; s. c. zhang , phys .
65 ( 1990 ) 120 .
j. m. p. carmelo , s. stlund , m. j. sampaio , ann .
( 2010 ) 1550 .
stellan " ostlund , eugene mele , phys .
b 44 ( 1991 ) 12413 . j. stein , j. stat .
88 ( 1997 ) 487
. f. kmpfer , m. moser , u .- j .
wiese , nucl .
b 729 ( 2005 ) 317 .
h. bethe , z. phys .
71 ( 1931 ) 205 .
e. k. sklyanin , l. a. takhtadzhan , l. d. faddeev , theor .
40 ( 1979 ) 194
. b. sriram shastry , j. stat .
50 , 57 ( 1988 ) .
a. damascelli , z. hussain , z .- x .
shen , rev .
phys . 75 ( 2003 ) 473 .
p. a. lee , n. nagaosa , x .-
wen , rev .
78 ( 2006 ) 17 .
stellan stlund , mats granath , phys .
96 ( 2006 ) 066404 .
t. senthil , ashvin vishwanath , leon balents , subir sachdev , matthew p. a. fisher , science 303 ( 2004 ) 1490 ; t. senthil , leon balents , subir sachdev , ashvin vishwanath , matthew p. a. fisher , phys . rev .
b 70 ( 2004 ) 144407 .
a. l. chernyshev , d. galanakis , p. phillips , a. v. rozhkov , a .- m .
s. tremblay , phys . rev .
b 70 ( 2004 ) 235111 .
m. takahashi , j. phys .
c 10 ( 1977 ) 1289 .
dionys baeriswyl , jos carmelo , kazumi maki , synth . met .
21 ( 1987 ) 271 .
shiping feng , z. b. su , l. yu , phys .
b 49 ( 1994 ) 2368 .
p. w. anderson , mater .
( 1973 ) 153 ; p. w. anderson , science 235 ( 1997 ) 1196 ; g. baskaran , z. zou , p. w. anderson , sol .
stat . commun .
63 ( 1987 ) 973 ; s. liang , b. douot , p. w. anderson , phys .
61 ( 1988 ) 365 .
wen , quantum field theory of many - body systems , oxford university press , oxford , 2004 .
masao ogata , hiroyuki shiba , phys .
b 41 ( 1990 ) 2326 ( 1990 ) .
f. woynarovich , j. phys .
c 15 ( 1982 ) 97 .
k. borejsza , n. dupuis , phys .
b 69 ( 2004 ) 085119 .
j. m. p. carmelo , j. m. romn , k. penc , nucl .
b 683 ( 2004 ) 387 .
m. veltman , facts and mysteries in elementary particle physics , world scientific , new jersey , 2003 .
l. pauling , proc .
london , ser .
a 1 , 96 ( 1949 ) 343 ; p. fazekas , p. w. anderson , philos . mag . 30
( 1974 ) 423 .
e. fradkin , field theories of condensed matter systems , addison - wesley , redwood city , california , 1991 .
a. auerbach , interacting electrons and quantum magnetism , springer - verlag , new york , 1998 .
n. m. r. peres , m. a. n. arajo , phys .
b 65 ( 2002 ) 132404 .
r. coldea , s. m. hayden , g. aeppli , t. g. perring , c. d. frost , t. e. mason , s .- w .
cheong , t. z. fisk , phys .
86 ( 2001 ) 5377 .
y. r. wang , phys .
b 46 ( 1992 ) 151 .
g. f. giuliani , g. vignale , quantum theory of the electron liquid , cambridge university press , cambridge , 2005 .
j. m. p. carmelo , t. prosen , d. k. campbell , phys .
b 63 ( 2001 ) 205114 ; j. m. p. carmelo , a. h. castro neto , d. k. campbell , phys . rev .
73 ( 1994 ) 926 ; 74 ( 1995 ) 3089 , erratum
. j. carmelo , d. baeriswyl , phys .
b 37 ( 1988 ) 7541 .
j. m. p. carmelo , p. d. sacramento , phys .
b 68 ( 2003 )
085104 . j. m. p. carmelo , k. penc , d. bozi , nucl .
b 725 ( 2005 ) 421 ; j. m. p. carmelo , k. penc , d. bozi , nucl . phys .
b 737 ( 2006 ) 351 , erratum .
j. m. p. carmelo , d. bozi , k. penc , j. phys .
: cond . mat . 20
( 2008 ) 415103 ; j. m. p. carmelo , k. penc , l. m. martelo , p. d. sacramento , j. m. b. lopes dos santos , r. claessen , m. sing , u. schwingenschlgl , europhys
67 ( 2004 ) 233 ; j. m. p. carmelo , p. horsch , d. k. campbell , a. h. castro neto , phys .
b ( rc ) 48 ( 1993 ) 4200 .
j. m. p. carmelo , k. penc , j. phys .
18 ( 2006 ) 2881 .
d. pines , p. nozires , in : the theory of quantum liquids , benjamin , new york , 1996 , vol | in this paper a general description for the hubbard model with nearest - neighbor transfer integral @xmath0 and on - site repulsion @xmath1 on a square lattice with @xmath2 sites is introduced .
it refers to three types of elementary objects whose occupancy configurations generate the state representations of the model extended global @xmath3 symmetry recently found in ref . @xcite .
such objects emerge from a suitable electron - rotated - electron unitary transformation .
it is such that rotated - electron single and double occupancy are good quantum numbers for @xmath4 .
the advantage of the description is that it accounts for the new found hidden @xmath5 symmetry in @xmath6/z_2 ^ 2 $ ] beyond the well - known @xmath7/z_2 $ ] model ( partial ) global symmetry .
specifically , the hidden @xmath5 symmetry state representations store full information on the positions of the spins of the rotated - electron singly occupied sites relative to the remaining sites . profiting from that complementary information , for the whole @xmath8 interaction range independent spin state representations
are naturally generated in terms of spin-@xmath9 spinon occupancy configurations in a spin effective lattice .
for all states such an effective lattice has as many sites as spinons .
this allows the extension to intermediate @xmath10 values of the usual large-@xmath10 descriptions of the spin degrees of freedom of the electrons that singly occupy sites , now in terms of the spins of the singly - occupied sites rotated electrons .
the operator description introduced in this paper brings about a more suitable scenario for handling the effects of hole doping . within it
such effects are accounted for in terms of the residual interactions of the elementary objects whose occupancy configurations generate the state representations of the charge hidden @xmath5 symmetry and spin @xmath11 symmetry , respectively .
this problem is investigated elsewhere .
the most interesting physical information revealed by the description refers to the model on the subspace generated by application of one- and two - electron operators onto zero - magnetization ground states .
( this is the square - lattice quantum liquid further studied in ref .
@xcite . )
however , to access such an information one must start from the general description introduced in this paper , which refers to the model in the full hilbert space . |
over the last several decades , a great deal of interest has been paid on the ( cumulative ) gravitational lensing effect on distant sources due to inhomogeneites in the matter distribution of the universe .
this problem has been studied using various methods @xcite . in some cases ,
the lens objects can be treated as point particles .
the point particles may be galaxies for cosmological lenses , or stars for microlensing events . while surveying these papers , one question arises to us out of purely theoretical interst : what happens when we bring the number of particles @xmath0 very large , while fixing the mass density ? does it look like a friedman
lematre(fl ) universe , or a completely different universe ? in this article , we discuss what the universe looks like when we take the large @xmath0 limit of a universe filled with point particles by studying the distance - redshift relation .
we distribute point particles of the same mass @xmath1 uniformly throughout the universe with a mean separation @xmath2 .
we assume that on large scales , the spacetime is described by an isotropic homogeneous metric ( robertson walker metric ) .
the energy density parameter @xmath3 is of order of @xmath4 .
condier a photon beam which is emitted from a distant source which we also treat as a point source .
we observe the redshift and the luminosity of this source . during the propagation
, the luminosity of the photon beam may be amplified by the gravinational lensing effect .
we can consider two types of lensing effect : * strong lensing effect ; when the beam passes very near to a point particle , it suffers a strong apmlification .
* cumulative weak lensing effect ; the beam does not pass very near to any particle , but travels through the `` ripples '' of gravitational potential , and suffers a weak amplification many times .
the cumulative amplification of the weak lensing effect is estimated as follws@xcite .
the expansion , @xmath5 , of the null geodesic satisfies @xmath6 where @xmath7 is the affine parameter and @xmath8 is the shear ; @xmath9 .
we neglected the vorticity .
since we assume that the beam does not pass very near the particles , the evolution of the shear @xmath8 is estimated as @xmath10 thus , the change in @xmath8 during passing by one particle is approximately given by @xmath11 then , the `` random walk '' for distance @xmath12 results in @xmath13.\end{aligned}\ ] ] this leads to @xmath14.\end{aligned}\ ] ] since we assume @xmath15 and @xmath16 where @xmath17 is the hubble constant , we obtain @xmath18 < < 1 , \label{eq : weak1}\\ \theta(l)l & \sim & { \cal o}[(l / l ) ] < < 1 .\label{eq : weak2}\end{aligned}\ ] ] in the fl case , on the contrary , the shear term vanishes and the ricci focusing term determines the evolution of the expansion ; @xmath19 this leads to @xmath20 .
thus , the cumulative amplification is negligible in the point mass universe when the number of paritcles is large enough .
note that we have assumed that the relation between the affine parameter and the redshift coincides with that of fl model .
the proof that the difference is negligible is given in @xcite .
next , we estimate the probability that the photon beam suffers the strong lensing effect as follows . we adopt the thin lens approximation .
the einstein radius of the lens whose mass is @xmath1 is given by@xcite @xmath21 where @xmath22 is the distance from the observer to the source .
kpc when @xmath23 , and @xmath24 times the solar radius when @xmath25 , where @xmath26 is the redshift of the lens .
thus , a star is usually smaller , and a galaxy is larger , than its einstein radius . ]
the magnification factor @xmath27 , the ratio of the flux density in the observed image to the flux density in the absence of the lens , is @xmath28 where @xmath29 is the impact parameter . , and becomes larger as @xmath29 increases . ] from this expression , we can tell that the source is strongly amplified ( i.e. , @xmath30 becomes of order of 10% ) when the impact parameter is close to the einstein radius ; @xmath31 .
we cam estimate the probability @xmath32 that the source is strongly amplified , by considering the probability of hitting any of @xmath0 discs of radius @xmath33 when a photon beam travels in a tube of area @xmath34 ( see fig.1 ) ; @xmath35 where @xmath12 is the distance to the source , and @xmath3 is the mean mass density of the point mass universe .
since @xmath36 and @xmath37 where @xmath17 is the hubble constant , the probability @xmath32 is much smaller than unity for sources at low redshifts , or in a low density universe , and @xmath38 even at rather high redshifts @xmath39 . when @xmath40 is smaller than unity , part of the beams @xmath41 will reach us without hitting any disc .
the distance ( which is estimated from the observed flux ) to such sources is obtained by following the evolution of the flux in an empty spacetime .
that is , we can regard the distance to these sources as so - called dyer - roeder distance@xcite .
therefore , when we observe the luminosity and redshift of distant point sources in the point mass universe , we would obtain the following : * part of sources @xmath41 follow the distance - redshift relation of dyer - roeder distance ; * other sources @xmath40 are strongly amplified . the fraction , @xmath41 , of the sources which never hit a dics is invariant when we change the value of @xmath0 . also , the statistical nature of the distribution of amplification factor is clearly independent of @xmath0 when @xmath40 is enough smaller than unity .
therefore , we can say * these natures are independent of the number of particles @xmath0 if the mass density @xmath3 is fixed .
actually , this is the well known fact `` the optical depth of gravitational lens is independent of the mass of the lens . ''
we here point out that the statement is valid even in the large @xmath0 limit .
that is , even if the universe is filled not with stars but with much smaller point particles , the distance - redshift relation of point sources satisfies the above features , as long as we can keep our assumptions , such as geoemtrical optics treatment .
the fraction @xmath40 increases for higher redshifts and a high density universe .
then , @xmath40 becomes larger than unity .
the number of beams which never hit a disc becomes very small , and multi - scattered events dominate .
however , we expect that the resulting distribution of luminosity of distant point sources is insensitive to @xmath0 as long as the distribution of point particles is random and the thin lens approximation holds .
this is because we can repeat the same discussion for the probability of suffering a next strong lensing event after hitting one disc .
the above analysis is valid for any matter as long as they are compact enough and interact with photons only through gravity .
the lensing objects need not to be galaxies or stars ; they may be sands or elementary ( but dark ) particles .
moreover , the thin lens approximation we adopted above seems to become better as we decrease the mass of lens , since the ratio of einstein radius @xmath33 to the means separation @xmath2 becomes smaller ; @xmath42 .
clearly , the effect of cumulative weak lensing effect becomes negligible as we can see from equations ( [ eq : weak1 ] ) and ( [ eq : weak2 ] ) .
thus , we conclude that the behaviour of distance - redshift relation of the point mas universe does not agree with that of a fl model even when we take the large @xmath0 limit .
holz and wald @xcite studied the lensing effect when the matter distribution of the universe is not homogeneous but the masses are concentrated into compact objects .
they commented that the probability distributions of lensing effect are indistinguishable between the cases where the masses of lenses are @xmath43 and @xmath44 .
their results may support the correctness of the above analysis . related with this
, one can show that , in their formalism , the observed ditance - redshift relation for any point source follows that of a fl model in the case of uniform ( not discrete ) distribution of matter , though they do not give an explicit statement .
we give a rough proof in the appendix . in sugiura
et al.@xcite , it was shown that the discreteness of matter distribution is harmless when we consider the distance - redshift relation in a spherically symmetric space .
it suggests that , in the point mass universe , if we average the luminosity of the sources of the same redshift over the whole sky and calculate the distace - redshift relation with the averaged luminosity , it should agree with the fl relation . holz and wald @xcite state that the averaged luminosity of the beams agrees with that of a fl model .
these statements justify the result we obtained in @xcite .
we also notice that , we would obtain the fl relation if we take the average of sources over a region larger than the mean separation of the interviening lens objects , i.e. , over the region which includes enough strong lensing events .
that is , when the source is much larger than the intervening lens objects .
we can safely calculate its distance using a fl model . + + the author would like to thank k. ioka and t. hamana for useful comments .
we start from the geodesic deviation equation .
let @xmath45 be the deviation vector , and define matrix @xmath46 by @xmath47 where @xmath48 is the affine parameter .
then the geodesic deviation equation is written as @xmath49 where @xmath50 is the tangent vector of the null geodesic . in a robertson - walker spacetime
, this equation takes the form @xmath51 where @xmath52 is the frequency of the photon and @xmath3 is the energy density of the universe .
then , after traveling for small @xmath53 , @xmath54 consider a ball of radius @xmath22 whose density is uniform , and a bundle of light ray which passes through this ball at a distance @xmath29 from the center of the ball . by direct calculation , they show that @xmath55 where @xmath56 where @xmath57 is the total mass of the matter inside the ball . from the relations @xmath58 and @xmath59
we can see that equation ( [ dev2 ] ) agrees with ( [ dev1 ] ) .
dyer and r.c .
roeder , astrophys .
j. * 174 * , l115 ( 1972 ) ; c.c .
dyer and r.c .
roeder , astrophys .
j. * 180 * , l31 ( 1973 ) n. sugiura , k. nakao , and t. harada , to appear in phys .
d * , ( 1998 ) , gr - qc/9808012 | we study the distance - redshift relation in a universe filled with point particles , and discuss what the universe looks like when we make the number of particles @xmath0 very large , while fixing the averaged mass density . using the raychaudhuri equation and a simple analysis of the probability of strong lensing effects ,
we show that the statistical nature of the amplification is independent of @xmath0 , and clarify the appearance of the point particle universe . 0h_0 @ @.2326ex # 1^to-1.4@-@@ + gravinational lens effect + norimasa sugiura , + _ department of physics , kyoto university , sakyo - ku , kyoto 606 - 8502 , japan _ |
low - dimensional quantum spin systems with frustrated interactions have drawn considerable attention for several decades @xcite .
in particular , the square - lattice @xmath8 frustrated heisenberg magnets with first - neighbor exchange constant @xmath9 and second - neighbor constant @xmath10 are increasingly interesting due to their unusual ground states and quantum phenomena @xcite-@xcite . based on the @xmath6 model studies , there exist several phases as a function of @xmath11 . when @xmath9 dominates or @xmath10 is ferromagnetic ( fm ) , the system is either nel antiferromagnetic ( nafm ) or fm depending on the sign of @xmath9 ( refs .
when @xmath10 is antiferromagnetic ( afm ) and dominates , there appears the so - called columnar afm ( cafm ) order @xcite with antiferromagnetically coupled fm chains .
the cafm and fm , or cafm and nafm ordered phases are separated by the intermediate quantum - disordered phases , the nature of which is not yet fully resolved @xcite .
the recent discoveries of quasi - two - dimensional materials are realizations to test the validity of the @xmath6 model .
prominent among them are li@xmath1vo(si , ge)o@xmath12 ( refs .
@xcite ) , @xmath13(vo)(po@xmath12)@xmath1 ( @xmath14=pb , zr , sr , ba ) ( refs .
@xcite ) , ( cubr)@xmath15o@xmath16 ( @xmath17=ca , sr , ba , pb ; @xmath18=nb , ta ) ( ref .
@xcite ) , and ( cu@xmath0)lanb@xmath1o@xmath2 ( @xmath0=cl , br ) ( refs .
( cu@xmath0)lanb@xmath1o@xmath2 compounds are of particular interest because they allow systematic tuning and understanding of the structural and magnetic properties , which are plausibly connected with the phenomenon of high-@xmath19 superconducting cuprates .
although divalent copper with the electronic configuration @xmath20 should be jahn - teller active and lead to the cooperative lattice distortion ( e.g. , perovskite kcuf@xmath21 ( ref .
@xcite ) ) , the precise crystal structure of the layered copper oxyhalides ( cu@xmath0)lanb@xmath1o@xmath2 is still under debate .
earlier structural studies on ( cu@xmath0)lanb@xmath1o@xmath2 were carried out with the tetragonal space group @xmath22 , where the cu and @xmath0 sites possess the @xmath23 symmetry @xcite . while the rietveld refinement gave satisfactory results , the thermal parameter for halogens remained large .
besides , in this structure copper is in a significantly squeezed octahedral coordination with two short cu - o bonds ( about 1.9 ) and four rather long cu-@xmath0 bonds ( 2.7 ) , which are also quite unusual .
subsequently , the neutron diffraction experiment @xcite proposed that the cl ions in ( cucl)lanb@xmath1o@xmath2 ( ( cucl)lno ) shifted away from the ideal wyckoff @xmath24 position @xcite . the transmission electron microscopy measurement on ( cucl)lno ( ref .
@xcite ) revealed superlattice reflections corresponding to an enlarged @xmath3 unit cell .
the nuclear magnetic resonance and the nuclear quadrupole resonance experiments for ( cucl)lno and ( cubr)lanb@xmath1o@xmath2 ( ( cubr)lno ) further demonstrated the lack of the tetragonal symmetry at both cu and cl / br sites @xcite .
the magnetic properties of ( cucl)lno and ( cubr)lno are also unusual and lack a clear microscopic interpretation .
the former exhibits a spin liquid phase with a spin gap @xcite that are incompatible @xcite with the square @xmath6 model . on the other hand ,
it has been reported @xcite that the replacement of cl by br leads to a cafm order in ( cubr)lno at low temperatures .
however , it is unclear whether the cu ions connected with the dominant exchange interaction couple ferromagnetically or antiferromagnetically @xcite . moreover , both ( cucl)lno and ( cubr)lno are claimed to be fm @xmath9 compounds @xcite whose justifications largely rely on the @xmath6 model . yet , the structural study @xcite raised doubts over the validity of the model
. therefore , unambiguous determination of the crystal structure is crucial for understanding these complex systems . at present , there are several structural models proposed for the cu@xmath0 plane .
whangbo and dai @xcite suggested a model that consists of different ring clusters to explore the exchange couplings .
however , the existence of inequivalent cu and cl sites in such a model is in contradiction to the experimental results that both cu and @xmath0 occupy a unique crystallographic site with no substantial disorder @xcite .
@xcite , based on the empirical evidence , proposed an orthorhombic distorted @xmath3 structure ( hereafter referred to as the yy model ) . in this model
, the displacement of cl ions generates different exchange couplings among the nearest neighboring cu pairs .
a cu dimer formed by the dominant exchange interaction was considered @xcite to study the spin - gap behavior .
the same structural model was shown @xcite to consistently account for ( cubr)lno . the third model , suggested by tsirlin and rosner ( tr ) @xcite is also characterized by an ordering pattern but with a @xmath25 periodicity , where the local environment of copper is distorted to form the cuo@xmath1cl@xmath1 plaquette .
first - principles calculations have proven to be an appealing method to deal with complex systems @xcite .
such a method can efficiently and reliably calculate the total energy , which is crucial in determining the most stable structure in order to study all relevant physical properties . in this work
, we will investigate the atomic structure and resultant magnetic property of ( cucl)lno and ( cubr)lno based on the density functional theory .
our results show that , among several dozens of examined structures , the distortion pattern of the most stable one is similar to that of the yy model .
the displacement of the @xmath0 ions changes the environment of copper to form the cuo@xmath1@xmath26 plaquette .
in addition , these two materials crystallize in the space group @xmath4 . the fm chains in cafm (
cubr)lno are found to be along the direction which is contrary to the previous conjecture @xcite .
it will be shown that ( cucl)lno still belongs to the afm @xmath9 compound .
the first- and second - neighbor exchange couplings of ( cucl)lno and ( cubr)lno are also discussed in detail .
figure [ fig1 ] illustrates the basic crystal structure of the copper oxyhalides ( cu@xmath0)lanb@xmath1o@xmath2 .
it is made up of copper - halogen planes and nonmagnetic double - perovskite lanb@xmath1o@xmath2 slabs .
the la ions are located at the 12-coordinate sites of the double - perovskite slabs .
the cu@xmath0 planes and the lanb@xmath1o@xmath2 slabs alternate along the @xmath27 direction such that the copper is six - fold coordinated , bridging between the apical o ions of nbo@xmath5 octahedra and surrounded by four @xmath0 halogens .
because of the short cu - o bond length ( @xmath281.9 ) , the cu@xmath0 plane is more appropriately considered as a cu@xmath0o@xmath1 layer .
the initial structural study on ( cu@xmath0)lanb@xmath1o@xmath2 was carried out with the space group @xmath22 , where both cu and @xmath0 have the @xmath23 symmetry @xcite . in this model ( hereafter referred to as c4 ) , the cu and @xmath0 ions are located at the wyckoff @xmath29 and @xmath24 positions , respectively ( fig .
[ fig2](a ) ) .
later studies @xcite proposed that cl ions are displaced from the @xmath23-symmetry positions .
the yy @xmath30 model is represented in fig . [
fig2](b ) .
the displacement of @xmath0 ions on the cu@xmath0 plane leads to the formation of the @xmath0-cu-@xmath0-cu-@xmath0 zigzag chains , as indicated in fig . [
fig2](b ) .
the original equivalent and perpendicular cu chains are now distinguishable . here
, the direction extending along the zigzag chains is defined as the @xmath31 axis .
the present calculations were based on the generalized gradient approximation ( gga ) @xcite to the exchange - correlation energy functional of the density functional theory .
it is known @xcite that cu - derived oxide compounds are usually strongly correlated systems .
the correlation effect is important for the present systems to understand their ground state .
therefore , the on - site coulomb interaction @xmath32 for cu @xmath33 electrons was also included @xcite ( gga+@xmath32 ) in this work . since the on - site exchange interaction @xmath34 is expected to be less influenced by the solid state effects @xcite , the relation @xmath35 was used @xcite for different choices of @xmath32 .
the projector - augmented - wave potentials , as implemented in vasp @xcite , were employed for the interactions between the ions and valence electrons . the plane - wave basis set with an
energy cut - off of 500 ev was used . to minimize numerical uncertainties ,
structural optimizations were performed using a @xmath36 supercell for all the test structures unless specified otherwise .
the @xmath37 monkhorst - pack grids were taken to sample the corresponding brillouin zone .
the lattice parameters and atomic positions were relaxed until the total energy changed by less than @xmath38 ev per conventional cell and the residual force was smaller than 0.01 ev / .
we first calculated the total energies of the c4 structure and several @xmath39 and @xmath36 distorted structures with different displacement patterns of the @xmath0 halogens .
the yy @xmath36 model is found to be the most stable one . as compared to the c4 structure
, the yy model has a significant 0.3 and 0.2 ev / fu lowering in the energy of ( cucl)lno and ( cubr)lno , respectively .
this directly rules out the possibility that the two compounds crystallize in the @xmath23 symmetry .
particularly , over the full @xmath32 range from 0 ev to 8 ev , the yy model is 0.1 ev / fu lower than the tr @xmath39 model ( see fig . 4 in ref .
@xcite ) for both materials . the nonmagnetic calculations with and without @xmath32 lead to the same conclusion . therefore , _ the structural distortion outweights the magnetism and on - site correlation effects _ in determining the atomic structure .
note that , besides refs .
@xcite and @xcite , very recent experimental evidences @xcite also confirm that the original unit cell should be double along both the @xmath40 and @xmath31 axes for the family of these compounds .
our study therefore provided theoretical support for the stabilization of the yy @xmath36 model . to examine whether there exists other more stable structure with the @xmath0 ions restricted to the cu plane
, we perform the calculations for ( cucl)lno with twenty sets of random displacements of all four cl ions from the positions in the yy model .
however , no such structure was found .
the resultant configurations of the trial structures are either relaxed back to the yy model or trapped into a nearby higher energy minimum .
next , we allow the halogens in the yy model to move off the plane .
it is found that the @xmath0 ions in the relaxed structures are @xmath41 away from the cu plane . however , the change in the total energy is rather small . at @xmath32=0 and 8 ev ,
the results of both materials show that the energy differences are only within 2 mev per @xmath36 supercell while the energies for the structures with the @xmath0 ions fixed in the plane remain lower .
we also examine the two structures in figs .
16(e ) and 16(f ) of ref .
@xcite , which are based on another @xmath3 configuration with the cl ions displaced away from the cu plane in a different way .
the calculations indicate that , after relaxation , both are energetically about 0.2 ev / fu higher than the yy model .
the increase in the total energy mostly comes from the different in - plane structural distortions .
again , the contribution from the @xmath42-component shift of cl ions is rather minor .
hence , the distortion on the cu@xmath0 plane is predominantly crucial to stabilize the atomic structure . in the following discussion
, we shall focus on the yy model with @xmath0 ions kept in the cu plane @xcite .
now , we analyze the total energies influenced by the on - site coulomb interaction and the different magnetic configurations shown in fig .
[ fig3 ] . here ,
sc1 and sc3 are fm and nafm .
sc2 and sc4 are both cafm , with the fm chain along the @xmath31 and @xmath40 directions , respectively .
the results are displayed in fig .
[ fig4 ] , where the energy of sc2 was chosen as a reference . for ( cucl)lno
, the energies of sc1 , sc2 , and sc4 are very competing .
the differences among them are within 1 mev / fu when @xmath43 .
the first two are even almost identical around @xmath32=4 ev . clearly , fig .
[ fig4](a ) shows that the sc3 is the lowest energy state and its energy is well separated from those of the other three magnetic structures . in the ( cubr)lno case , similar tiny energy differences but between sc1 , sc2 , and sc3 are also found .
interestingly , when the fm chain in the cafm state is set parallel along the @xmath40 axis , as in sc4 , the total energy over the examined @xmath32 range is much higher than those of the other three configurations , indicating that the cu ions along the @xmath40-axis should not couple ferromagnetically .
this finding is contrary to the previous conjecture @xcite .
the different energy ordering for the four magnetic configurations between the two compounds are conceivable since the magnetic interactions through the path cu-@xmath0-cu depend subtly on the small structural variation via the @xmath0-ion size effect .
we will return to this issue in sec .
iiid when considering the various exchange couplings .
tables [ tab1 ] lists the fully optimized structural parameters of both materials . for comparison , those obtained by the c4 model
are also included .
as can be seen in this table , the evaluated lattice constants are in good agreement with the experimental data @xcite .
the discrepancies between them are only within @xmath44 , the typical errors in the density - functional calculation .
the @xmath40 and @xmath31 lattice constants of ( cucl)lno are smaller than those of ( cubr)lno , which is due primarily to the size effect of br in the layered structure . to discuss the structural distortion ,
we take the ( cucl)lno case as an example . in the c4 model , copper is in the squeezed octahedral coordination with four long cu - cl bonds [ @xmath45(cu - cl)=2.77 ] and two short cu - o bonds [ @xmath45(cu - o)=1.85 ] .
the displacements of the cl and cu ions in the yy model lead to two shorter cu - cl bonds of 2.38 and 2.39 , forming the cl - cu - cl - cu - cl zigzag chain to stabilize the structure .
the rest two cu - cl interatomic distances are increased to @xmath46 . in particular ,
the cu - o bond length remains short after the structural distortion ( from 1.85 to 1.88 ) , indicating the strong bonding character between cu and o ions .
the calculated interatomic distances are comparable to those reported previously @xcite . as a result
, the distortion yields the nearly planar cuo@xmath1cl@xmath1 rather than the octahedral cuo@xmath1cl@xmath12 environment around the cu ion ( fig . [ fig5](a ) ) .
the resultant cuo@xmath1cl@xmath1 planar structure is reminiscent of the conventional cuo@xmath12 plaquette , which is commonly observed in copper oxides , e.g. , la@xmath1cuo@xmath12 ( ref .
@xcite ) and sr@xmath1cuo@xmath47 ( ref .
it should be noted that the cuo@xmath1cl@xmath1-plaquette zigzag chains was also reported in the tr model @xcite .
additionally , the basic electronic structure is similar to that of the cuo@xmath12 planar unit , which will be demonstrated in the next section .
combined with the energetic advantage mentioned above , the yy model provides a realistic description for the atomic structures of ( cucl)lno and ( cubr)lno . from a closer analysis of the positions of all ions in ( cucl)lno
, we found that the distorted atomic structure in the yy model belongs to the space group _ pbam _ ( no .
55 ) @xcite .
the atomic positions are summarized in table [ tab2 ] . clearly , the deviations of the cl ions from the @xmath23-symmetry positions are as large as 0.66 , and these values are four times larger than those of cu ions .
note that the displacements along the @xmath40 axis are more significant than those along the @xmath31 axis for both ions to form the zigzag chains . as expected , the structural distortion on the cucl plane leads la , nb and o ions to shift from the the @xmath23-symmetry positions .
figure [ fig5](b ) and table [ tab2 ] show the significant tilting and distortion of nbo@xmath5 octahedra .
such a tilting distortion is typical for perovskite oxides structures @xcite . particularly , la ions shift along the @xmath31 axis by an amount of 0.10 .
this displacement of la from the @xmath23-symmetry positions well agrees with the experimental nonzero value of the efg tensor at la sites @xcite , a strong evidence for the structural distortion in ( cucl)lno .
we also found that nb ions shift along the @xmath40 axis by a relatively smaller amount of 0.02 . note that the @xmath40 ( @xmath31 ) component of la ( nb ) displacement is almost negligible .
taking into account the tilting of the nbo@xmath5 octahedra in the ( cucl)lno is important for providing a realistic description of the distortion on the cucl plane .
figure [ fig5](c ) , the top view of the atomic structure , clearly demonstrates that the cooperative tilting of the nbo@xmath5 octahedra in the space group @xmath4 results in a @xmath3 periodicity and leads to the zigzag chains with the same periodicity .
it is worth pointing out that the higher symmetric @xmath25 zigzag chains in the tr model were investigated without consideration of the effect due to the tilting of the nbo@xmath5 octahedra , where these octahedra were still kept at the c4 tetragonal sites @xcite . allowing the tilting distortion of nbo@xmath5 octahedra in ( cucl)lno lowers the symmetry of the atomic structure and correspondingly that of the zigzag chains and therefore leads to a lower total energy . in the yy model ,
the zigzag chains have the glide symmetry about @xmath48 and @xmath49 .
specifically , the cu - cl bond of 2.39 in the zigzag chain is next to the cl - cu bond of 2.38 in the adjacent chain and vice versa .
as compared to those in the tr model , such a _ complementary _ arrangement between adjacent chains in the yy model allows a further lowering in energy .
now , it is evident that the yy @xmath3 model in the present study is energetically more stable than the tr @xmath25 one .
similar conclusion holds for ( cubr)lno .
figure [ fig6 ] depicts the orbital- and site - projected densities of states ( dos ) of ( cucl)lno with @xmath32=6 ev , where the valence - band maximum ( @xmath50 ) is set to zero .
the orbitals are projected in the local coordinates with the @xmath51 and @xmath52 axes directed to the neighboring cl ions and the @xmath42 axis coinciding with the crystal @xmath27 axis ( fig .
[ fig2](b ) ) . among the major valence state region of 6.3 ev ,
the higher - energy part consists almost exclusively of o and cl @xmath53 states .
there is larger contribution from the cl @xmath53 state just below @xmath50 .
the lower - energy part , dominated by the cu @xmath45 states , is splitted into the doubly occupied @xmath54 , @xmath55 , @xmath56 , @xmath57 , and singly occupied @xmath58 states .
as compared to the gga dos ( not shown here ) , the gga+@xmath32 shows an essential redistribution of the cu @xmath33 dos , i.e. , from being above to below the o and cl @xmath53 states . that the energy gap ( @xmath59 ) lies between the occupied anion @xmath53 states and the unoccupied cu @xmath45 states is similar to those in the charge - transfer insulators , e.g. , la@xmath1cuo@xmath12 ( ref .
@xcite ) and sr@xmath1cuo@xmath47 ( ref .
the sharp peak of the low - lying cu @xmath58 state is a result of the strong bonding between the cu and o ions with a considerably short cu - o bond length of 1.88 ( see table [ tab1 ] ) .
note that the @xmath58 orbital was hybridized with little cl @xmath53 component .
the on - site correlation @xmath32 leads to the half - filling of the @xmath58 orbital , of which the lobes point to the o ions .
these results imply a single orbital ground state .
figure [ fig6 ] shows that , due to the hybridization with the o @xmath53 state @xcite , the @xmath58 bonding - antibonding separation ( 8.3 ev ) is larger than the value of @xmath32 .
therefore , the electronic structure due to the cuo@xmath1cl@xmath1 plaquette in the yy model is very similar to those of other copper oxides @xcite with planar cuo@xmath12 units .
we found that the structural distortion and magnetism together already open up the gap .
the @xmath59 of ( cucl)lno obtained by the gga is 0.3 ev .
however , this result is insufficient to account for the green color appearance @xcite of this compound . at @xmath60 ev , the @xmath59 is increased to 1.8 ev .
further increase of @xmath32 makes no significant widening for the band gap .
the main structures in the dos of ( cucl)lno are also found in that of ( cubr)lno , except for the smaller @xmath59 of 1.5 ev . at this choice of @xmath32 , the local magnetic moment at the cu site of (
cubr)lno is evaluated to be 0.6 bohr magneton , which agrees with the experiments @xcite .
an amount of 0.1 bohr magneton at br sites is also observed .
hence , we choose the optimal @xmath32 = 6 ev case to discuss the corresponding atomic and electronic properties . finally , we discuss the exchange couplings for both ( cucl)lno and ( cubr)lno . in the undistorted c4 structure , the interactions between the cu ions can be approximately modeled by the heisenberg hamiltonian @xmath61 , where the sums run over the first and second nearest - neighbor pairs , respectively , and @xmath62 corresponds to the spin moment at site @xmath63 .
the relevant exchange couplings can be then determined by applying the model to the calculated energies of different spin configurations . for the yy model ,
the formation of the @xmath0-cu-@xmath0-cu-@xmath0 zigzag chains along the @xmath31 axis ( fig . [ fig2](b ) ) lifts the tetragonal symmetry and leads to inequivalent superexchange pathways , as indicated in fig .
[ fig2 ] . as a result
, the @xmath9 in the c4 structure is split into @xmath64 , @xmath65 , and @xmath66 , with the former two now being the first neighboring inter - chain interactions and the latter the first neighboring intra - chain interaction .
the original @xmath10 coupling is split into two inequivalent @xmath67 and @xmath68 , which are correspondingly the second neighboring inter - chain interactions .
we investigate these interactions via the various spin configurations in fig .
the results are summarized in table [ tab3 ] .
we first discuss the results from the c4 model .
table [ tab3 ] shows that @xmath9 is almost negligible as compared to @xmath10 .
this is reasonable because , as illustrated in fig .
[ fig7](a ) , there is no overlap between the cu2 @xmath57 and @xmath04 @xmath53 orbitals .
therefore , even with the obvious overlapping of the cu1 @xmath45 and @xmath04 @xmath53 orbitals , cu1 and cu2 could hardly interact with each other . on the other hand , cu1 can interact with cu3 via the @xmath04 @xmath53 orbital .
based on the @xmath6 model , both ( cucl)lno and ( cubr)lno were previously claimed @xcite to be fm @xmath9 magnets with competing afm @xmath10 interactions , as in the case of pb@xmath1vo(po@xmath12)@xmath1 ( ref .
@xcite ) . table [ tab3 ] indeed shows that @xmath69 and @xmath70 for both materials in the c4 model , a direct consequence of the hund s coupling and virtual electron hopping .
however , the recent structural study @xcite has raised serious doubt over the validity of the @xmath6 model in such materials .
our calculations also indicate that consideration of the structural distortion leads to the opposite results . for ( cucl)lno ,
the effective interactions @xmath71 and @xmath72 in the yy model are found to be afm and fm , respectively .
and they both become fm for ( cubr)lno at large @xmath32s .
these results come from the complicated interplay between the hund s coupling , virtual electron hopping , the distorted structure and @xmath0-ion size effect .
it should be noted that the tr model @xcite also results in a leading afm coupling in ( cucl)lno .
this implies that the simple @xmath6 model is unable to describe the present systems .
moreover , the first neighboring interactions become more significant as compared to the second neighboring ones .
figure [ fig7](b ) clearly shows that , unlike the c4 case , @xmath73 cu1-@xmath74-cu2 is no longer 90@xmath75 ( table [ tab1 ] ) due to the structural distortion .
this will lead to the overlapping of cu2 ( cu4 ) @xmath45 and @xmath04 @xmath53 orbitals , and enhance the interaction between cu1 and cu2 ( cu4 ) .
in fact , fig .
[ fig8](a ) shows that for ( cucl)lno , @xmath76 , @xmath77 , and @xmath78 for all the @xmath32s considered .
it is now clear that , since the interactions due to all the corresponding spin pairs in sc3 satisfy these conditions , ( cucl)lno in sc3 is much more stable than in the rest three configurations of fig .
[ fig3 ] . actually , sc3 is the most stable structure among all the spin configurations with the interactions up to second - nearest neighbors .
however , by comparing fig .
[ fig8](b ) with fig .
[ fig8](a ) , we found that the @xmath66 in ( cubr)lno becomes fm .
this could be ascribed to the fact that the interaction @xmath66 depends sensitively on the angle of the cu1-@xmath04-cu2 superexchange path in fig .
[ fig7 ] , and the replacement of cl by br will change this angle and modify the interaction . therefore , for ( cubr)lno , the first neighboring couplings within the chains are fm , and the first two neighboring couplings between adjacent chains are afm and fm , respectively . none of the four structures in fig .
[ fig3 ] satisfies these conditions .
specifically , the interactions due to the corresponding spin pairs in sc4 are all opposite to these couplings , giving rise to the result of the sc4 being the highest - energy structure for ( cubr)lno .
the above analysis seems to indicate that , contrary to previous expectations @xcite , ( cucl)lno rather than ( cubr)lno is less frustrated @xcite . to check the reliability of our calculations , we perform the total - energy calculation for an additional structure sc5 with three of the four spins being the same but opposite to the fourth one . for all the possible choices of four magnetic structures containing sc5 in solving the heisenberg hamiltonian , the deviations of the relevant couplings ( dashed lines in fig .
[ fig8 ] ) from those obtained by sc1@xmath79sc4 are less than 1.0 mev .
more importantly , the signs of these interactions remain unaltered for both materials . one plausible explanation for the above puzzling discrepancy is that the third neighboring couplings between different zigzag chains @xcite may not be completely negligible @xcite .
actually , in the yy model , the two cuo@xmath1@xmath26 plaquettes with the cu1 and cu5 ions in fig . [ fig7](b ) could be considered approximately coplanar .
@xcite argue that this kind of coplanarity provides an opportunity for the interaction between cu1 and cu5 through the overlap of the cu1 @xmath80x4 @xmath815 @xmath82cu5 @xmath57 orbitals .
so , from the structural geometry point of view , such a long - range coupling ( 8.5 ) could be possible .
however , to examine and identify these couplings , one has to take into account additional eight inequivalent couplings and use a larger supercell whose corresponding calculations are very time - consuming and yet , likely , not accurate enough for the present purposes ( fig .
[ fig4 ] ) .
therefore , we will not address this issue presently .
further theoretical and experimental work is required to clarify this point .
it is worth pointing out that the first neighboring interactions within the chains and between the adjacent chains have the opposite signs for ( cubr)lno .
the spatial asymmetry of these results again demonstrates the inappropriateness of the square @xmath6 model for the br compound .
furthermore , for both materials , the couplings between the adjacent chains are very competitive to the intra - chain interactions , sharply contrary to the previous conclusion @xcite .
tsirlin and rosner have argued @xcite that in the tr model , where the basic structure element is also the cuo@xmath1cl@xmath1-plaquette zigzag chain , the large hopping runs along the chain and leads to the strongest interaction . according to their discussion , the inter - chain interaction is rather weak due to the long `` nonbonding '' cu - cl distance and the lack of the proper superexchange path . in the present study , despite the similar backbone in the yy model , the couplings between the adjacent chains are shown to be still substantial . as mentioned before
, the strength of exchange interactions between two spin sites should be determined by the overlap of orbitals rather than the distance between them .
the interactions between the cu ions from adjacent chains could be significant through the path mediated by the extended @xmath83 orbital of cl ions ( vs o@xmath84 ) and would be enhanced in the br case with the further extended @xmath85 orbital .
indeed , our calculations show that all the in - plane exchange couplings in ( cubr)lno are three times larger than those in ( cucl)lno
. for the inter - layer interaction @xmath86 , table [ tab3 ] shows that the @xmath86 is afm , in agreement with the experiment @xcite .
when compared to the in - plane interaction , the @xmath86 in ( cucl)lno is non - negligible , implying that some long - path ( 12 ) interaction between the cu ions is still cooperative .
the origin of this long - range coupling could be associated with the interaction through the cu @xmath58 orbital . as discussed in sec .
iiic , this orbital strongly overlaps with the o @xmath87 orbital
. the o @xmath87 orbitals further couple with nb @xmath88 orbitals .
therefore , the inter - plane coupling @xmath86 shall involve the cu - o - nb - o - nb - o - cu path . in ( cubr)lno ,
however , the coupling is found to be relatively less significant .
the strength of all the interactions interested here is decreased with increasing @xmath32 .
the evolution is expected since adding @xmath32 makes the wavefunctions more localized and the virtual electrons hopping less favorable .
in conclusion , we have investigated the atomic structure and magnetic property of the copper oxyhalides ( cu@xmath0)lanb@xmath1o@xmath2(@xmath0=cl and br ) based on the density functional theory .
the calculations show that , among the examined structures , the yy @xmath3 model proposed by yoshida _
@xcite has the lowest energy .
this model is significantly more stable than both the undistorted c4 structure and the tr @xmath25 model suggested recently by tsirlin and rosner @xcite .
the @xmath0 and cu ions in the yy model are displaced to form the @xmath0-cu-@xmath0-cu-@xmath0 zigzag chains and the local environment of copper is distorted to form a nearly cuo@xmath1@xmath26 plaquette .
we found that ( cu@xmath0)lanb@xmath1o@xmath2 crystallizes in the space group @xmath4 .
the cooperative tilting of the nbo@xmath5 octahedra leads to the lower symmetry of the zigzag chains with a @xmath3 periodicity .
with consideration of the on - site coulomb interaction , the yy model shows the single - orbital scenario typical for copper oxides and oxyhalides .
we concluded that ( cucl)lno is still the afm @xmath9 magnet with mixing fm @xmath10 interactions . for ( cucl)lno ,
the first neighboring interactions within the zigzag chains are afm , and the first two neighboring couplings between adjacent chains are afm and fm , respectively .
however , the replacement of cl by br modifies the first neighboring intra - chain interaction to be fm for ( cubr)lno . despite
the `` well''-separated zigzag chains in the yy model , the couplings between adjacent chains are comparable to those within the chain .
the opposite signs of the inter- and intra - chain interactions in ( cubr)lno reflect the spatial asymmetry and therefore the failure of the simple @xmath6 model for such material .
all the in - plane exchange couplings in ( cubr)lno are shown to be three times those in the cl counterpart .
it is found that the inter - plane interaction @xmath86 is afm , in agreement with the experiment @xcite .
the present study strongly suggests that the simple square @xmath6 model should be modified to explore the magnetic property of cu@xmath0lanb@xmath1o@xmath2 .
we hope the present calculations will shed light on the precise crystallographic determination and the magnetic properties of cu@xmath0lanb@xmath1o@xmath2 .
we are grateful to p. sindzingre for bringing these systems to our attention .
computer resources provided by the national center for high - performance computing are gratefully acknowledged .
this work was supported by the national science council and national center for theoretical sciences of taiwan .
99 , edited by h.t .
diep ( world scientific , singapore , 1994 ) ; _ frustrated spin systems _ , edited by h.t .
diep ( world scientific , singapore , 2004 ) .
e. dagotto and a. moreo , phys .
lett . * 63 * , 2148 ( 1989 ) .
p. chandra , p. coleman , and a. i. larkin , phys .
64 * , 88 ( 1990 ) .
r. f. bishop , d. j. j. farnell , and j. b. parkinson , phys .
b * 58 * , 6394 ( 1998 ) .
r. r. p. singh , z. weihong , c. j. hamer , and j. oitmaa , phys .
b * 60 * , 7278 ( 1999 ) .
l. capriotti , f. becca , a. parola , and s. sorella , phys .
* 87 * , 097201 ( 2001 ) .
o. p. sushkov , j. oitmaa , and z. weihong , phys .
b * 63 * , 104420 ( 2001 ) .
j. sirker , z. weihong , o. p. sushkov , and j. oitmaa , phys .
b * 73 * , 184420 ( 2006 ) .
d. schmalfu , r. darradi , j. richter , j. schulenburg , and d. ihle , phys .
lett . * 97 * , 157201 ( 2006 ) .
j. r. viana and j. r. de sousa , phys . rev .
b * 75 * , 052403 ( 2007 ) .
a. ohare , f. v. kusmartsev , and k. i. kugel , phys .
b * 79 * , 014439 ( 2009 )
. n. shannon , b. schmidt , k. penc , and p. thalmeier , eur .
j. b * 38 * , 599 ( 2004 ) .
n. shannon , t. momoi , and p. sindzingre , phys .
* 96 * , 027213 ( 2006 ) .
p. millet and c. satto , mater .
bull . * 33 * , 1339 ( 1998 ) .
r. melzi , p. carretta , a. lascialfari , m. mambrini , m. troyer , p. millet , and f. mila , phys .
85 * , 1318 ( 2000 ) .
e. kaul , ph.d .
thesis , technische universit@xmath89 t dresden , dresden , 2005 .
n. s. kini , e. e. kaul , and c. geibel , j. phys . : condens .
matter * 18 * , 1303 ( 2006 ) .
e. e. kaul , h. rosner , n. shannon , r. v. shpanchenko and c. geibel , j. magn .
magn . mater .
* 272 - 276(ii ) * , 922 ( 2004 )
. y. tsujimoto , h. kageyama , y. baba , a. kitada , t. yamamoto , y. narumi , k. kindo , m. nishi , j. p. carlo , a. a. aczel , t. j. williams , t. goko , g. m. luke , y. j. uemura , y. ueda , y. ajiro , and k. yoshimura , phys .
b * 78 * , 214410 ( 2008 )
. m. yoshida , n. ogata , m. takigawa , j. yamaura , m. ichihara , t. kitano , h. kageyama , y. ajiro , k. yoshimura , j. phys .
. jpn . * 76 * , 104703 ( 2007 ) .
m. yoshida , n. ogata , m. takigawa , t. kitano , h. kageyama , y. ajiro , k. yoshimura , j. phys .
* 77 * , 104705 ( 2008 ) .
a. i. liechtenstein , v. i. anisimov , and j. zaanen , phys .
b * 52 * , r5467 ( 1995 ) .
t. a. kodenkandath , j. n. lalena , w. l. zhou , e. e. carpenter , c. sangregorio , a. u. falster , w. b. simmons , c. j. oconnor , j. b. wiley , j. am .
soc . * 121 * , 10743 ( 1999 ) .
t. a. kodenkandath , a. s. kumbhar , w. l. zhou , and j. b. wiley , inorg . chem . * 40 * , 710 ( 2001 ) .
g. caruntu , t. a. kodenkandath , j. b. wiley , mater .
bull . * 37 * , 593 ( 2002 ) .
h. kageyama , j. yasuda , t. kitano , k. totsuka , y. narumi , m. hagiwara , k. kindo , y. baba , n. oba , y. ajiro , and k. yoshimura , j. phys .
. jpn . * 74 * , 3155 ( 2005 ) .
h. kageyama , t. kitano , n. oba , m. nishi , s. nagai , k. hirota , l. viciu , j. b. wiley , j. yasuda , y. baba , y. ajiro , and k. yoshimura , j. phys .
74 * , 1702 ( 2005 ) .
m. h. whangbo and d. dai , inorg .
45 * , 6227 ( 2006 ) .
n. oba , h. kageyama , t. kitano , j. yasuda , y. baba , m. nishi , k. hirota , y. narumi , m. hagiwara , k. kindo , t. saito , y. ajiro , and k. yoshimura , j. phys .
. jpn . * 75 * , 113601 ( 2006 ) .
a. a. tsirlin and h. rosner , phys . rev .
b * 79*,214416 ( 2009 ) . s. c. erwin , phys .
lett . * 91 * , 206101 ( 2003 ) .
a. malashevich and d. vanderbilt , phys .
* 101 * , 037210 ( 2008 ) .
h. rosner , r. r. p. singh , w. h. zheng , j. oitmaa , s. -l .
drechsler , and w. e. pickett , phys .
lett . * 88 * , 186405 ( 2002 ) .
j. p. perdew , k. burke , and m. ernzerhof , phys .
77 * , 3865 ( 1996 ) .
f. c. zhang and t. m. rice , phys .
b * 37 * , 3759 ( 1988 ) . m. t. czyyk and g. a. sawatzky , phys . rev .
b * 49 * , 14211 ( 1994 ) .
w. c. mackrodt and h. j. gotsis , phys .
b * 62 * , 10728 ( 2000 ) .
i. solovyev , n. hamada , and k. terakura , phys .
b * 53 * , 7158 ( 1996 ) .
g. kresse and d. joubert , phys . rev .
b * 59 * , 1758 ( 1999 ) .
g. kresse and j. furthmller , comput .
* 6 * , 15 ( 1996 ) .
a. kitada , y. tsujimoto , h. kageyama , y. ajiro , m. nishi , y. narumi , k. kindo , m. ichihara , y. ueda , y. j. uemura , and k. yoshimura , phys .
b * 80 * , 174409 ( 2009 ) . the model in ref . @xcite
originally proposed for ( fecl)lanb@xmath1o@xmath2 ( the same shown in fig .
16(a ) of ref .
@xcite ) is energetically less favorable than the yy model
. n. oba , h. kageyamaa , t. saito , m. azuma , w. paulus , t. kitano , y. ajiro , and k. yoshimur , j. magn .
* 310 * , 1337 ( 2007 ) . , edited by t. hahn ( kluwer , dordrecht , 1989 ) , vol .
a. , edited by c. n. r. rao and b. raveau ( world scientific , singapore , 1998 ) . from the viewpoint of the square @xmath6 model , these interactions should be regarded as the fourth - nearest - neighbor interactions .
h. kageyama , j. kang , c. lee , and m .- h .
whangbo , ( unpublished ) .
fig . 1 : ( color online ) crystal structure of ( cu@xmath0)lanb@xmath1o@xmath2 ( @xmath0=cl , br ) in the tetragonal space group @xmath22 . + + fig . 2 : ( color online ) ( a ) the undistorted tetragonal model and ( b ) the distorted model proposed by yoshida _
@xcite for ( cu@xmath0)lanb@xmath1o@xmath2 ( @xmath0=cl , br ) .
large and small spheres denote @xmath0 and cu ions , respectively .
the relevant exchange couplings are also indicated .
+ + fig .
3 : ( color online ) the four different spin configurations of cu ions considered in the present study . large and small spheres denote cl / br and cu ions , respectively .
+ + fig . 4 : ( color online ) total energies of ( a ) ( cucl)lanb@xmath1o@xmath2 and ( b ) ( cubr)lanb@xmath1o@xmath2 in the spin configurations shown in fig .
the energies are relative to that of sc2 .
+ + fig .
5 : ( color online ) perspective view of ( a ) the cuo@xmath90-plaquette zigzag chains and ( b ) tilted nbo@xmath5 octahedra of ( cu@xmath0)lanb@xmath1o@xmath2 ( @xmath0=cl , br ) in the space group @xmath4 .
( c ) top view of ( b ) . here ,
the symbols for the various atomic species are the same as those in fig .
+ + fig . 6 : ( color online ) orbital- and site - projected density of states ( dos ) of ( cucl)lanb@xmath1o@xmath2 , obtained by @xmath60 ev and sc2 in fig .
the energy is relative to the valence - band maximum .
+ + fig . 7 : ( color online ) schematic plot of cu @xmath57 and @xmath0 @xmath53 orbitals of ( cu@xmath0)lanb@xmath1o@xmath2 ( @xmath0=cl , br ) in the space group ( a ) @xmath22 and ( b ) @xmath4 .
+ + fig . 8 : ( color online ) the @xmath32-dependence of the in - plane exchange couplings of ( a ) ( cucl)lanb@xmath1o@xmath2 and ( b ) ( cubr)lanb@xmath1o@xmath2 .
the solid lines are calculated from sc1@xmath79sc4 in fig .
the dashed lines show the uncertainty of the calculations .
see the text for details .
note that the scale of exchange coupling in ( a ) is only half of that in ( b ) .
.the calculated lattice constants ( ) , relevant interatomic distances ( ) , and bond angles ( @xmath91 ) of ( cu@xmath0)lanb@xmath1o@xmath2 ( @xmath0=cl , br ) , obtained by @xmath60 ev and sc2 in fig .
the latter two refer to those in fig . [ fig2](b ) . in the bottom part
, the results in the undistorted tetragonal model are listed for comparison .
the values in parentheses are the corresponding experimental data . [ cols="<,<,<",options="header " , ] | the atomic and magnetic structures of ( cu@xmath0)lanb@xmath1o@xmath2 ( @xmath0=cl and br ) are investigated using the density - functional calculations . among several dozens of examined structures , an orthorhombic distorted @xmath3 structure , in which the displacement pattern of @xmath0 halogens resembles the model conjectured previously based on the empirical information is identified as the most stable one .
the displacements of @xmath0 halogens , together with those of cu ions , result in the formation of @xmath0-cu-@xmath0-cu-@xmath0 zigzag chains in the two materials .
detailed analyses of the atomic structures predict that ( cu@xmath0)lanb@xmath1o@xmath2 crystallizes in the space group @xmath4 .
the nearest - neighbor interactions within the zigzag chains are determined to be antiferromagnetic ( afm ) for ( cucl)lanb@xmath1o@xmath2 but ferromagnetic ( fm ) for ( cubr)lanb@xmath1o@xmath2 . on the other hand , the first two neighboring interactions between the cu cations from adjacent chains are found to be afm and fm respectively for both compounds .
the magnitudes of all these in - plane exchange couplings in ( cubr)lanb@xmath1o@xmath2 are evaluated to be about three times those in ( cucl)lanb@xmath1o@xmath2 . in addition , a sizable afm inter - plane interaction is found between the cu ions separated by two nbo@xmath5 octahedra .
the fourth - neighbor interactions are also discussed .
the present study strongly suggests the necessity to go beyond the square @xmath6 model in order to correctly account for the magnetic property of ( cu@xmath7lanb@xmath1o@xmath2 . +
pacs : 71.15.mb , 75.45.+j + |
the recent literature offers several results on the properties of gas flows on networks . for instance , in @xcite the well posedness is established for the gas flow at a junction of @xmath6 pipes with constant diameters .
the equations governing the gas flow in a pipe with a smooth varying cross section @xmath7 are given by ( see for instance @xcite ) : @xmath8 the well posedness of this system is covered in @xcite where an attractive unified approach to the existence and uniqueness theory for quasilinear strictly hyperbolic systems of balance laws is proposed .
the case of discontinuous cross sections is considered in the literature inserting a junction with suitable coupling conditions at the junction , see for example @xcite .
one way to obtain coupling conditions at the point of discontinuity of the cross section @xmath9 is to take the limit of a sequence of lipschitz continuous cross sections @xmath10 converging to @xmath9 in @xmath11 ( for a different approach see for instance @xcite ) .
unfortunately the results in @xcite require @xmath5 bounds on the source term and well posedness is proved on a domain depending on this @xmath5 bound . since in the previous equations the source term contains the derivative of the cross sectional area one can not hope to take the limit @xmath12 . indeed
when @xmath9 is discontinuous , the @xmath5 norm of @xmath13 goes to infinity .
therefore the purpose of this paper is to establish the result in @xcite without requiring the @xmath5 bound .
more precisely , we consider the cauchy problem for the following @xmath14 system of equations @xmath15 endowed with a ( suitably small ) initial data @xmath16 belonging to @xmath17 , the space of integrable functions with bounded total variation ( tot.var . ) in the sense of @xcite . here
@xmath18 is the vector of unknowns , @xmath19 denotes the fluxes , _
i.e. _ a smooth function defined on @xmath20 which is an open neighborhood of the origin in @xmath21 .
the system ( [ integrable : eq : bal ] ) is supposed to be strictly hyperbolic , with each characteristic field either genuinely nonlinear or linearly degenerate in the sense of lax @xcite . concerning the source term @xmath22
, we assume that it satisfies the following caratheodory type conditions : * @xmath23 is measurable with respect to ( w.r.t . )
@xmath24 , for any @xmath25 , and is @xmath26 w.r.t .
@xmath27 , for any @xmath28 ; * there exists a @xmath11 function @xmath29 such that @xmath30 ; [ hyp_on_g ] * there exists a function @xmath31 such that @xmath32 . note
that the @xmath11 norm of @xmath29 does not have to be small but only bounded differently from @xmath33 whose norm has to be small ( see theorem [ th : introduction ] below ) .
furthermore condition @xmath34 replaces the @xmath5 bound of the @xmath26 norm of @xmath22 in @xcite .
finally observe that we do not require any @xmath5 bound on @xmath35 .
on the other hand we will need the following observation : if we define @xmath36 by absolute continuity one has @xmath37 as @xmath38 .
moreover , we assume that a _ non - resonance _
condition holds , that is the characteristic speeds of the system ( [ integrable : eq : bal ] ) are bounded away from zero : | _ i(u)| c > 0 , u , i\{1, ,n}. [ hyp : non_resonance ] the following theorem states the well posedness of ( [ integrable : eq : bal ] ) in the above defined setting .
[ th : introduction ] assume @xmath39@xmath40 and ( [ hyp : non_resonance ] ) .
if the norm of @xmath35 in @xmath41 is sufficiently small , there exist a constant @xmath42 , a closed domain @xmath43 of integrable functions with small total variation and a unique semigroup @xmath44 satisfying a. @xmath45 for all @xmath46 and @xmath47 ; b. @xmath48 for all @xmath46 and @xmath47 ; c. for all @xmath49 the function @xmath50 is a weak entropy solution of the cauchy problem ( [ integrable : eq : bal])([init : data ] ) and satisfies the integral estimates ( [ eq : first_int_cond ] ) , ( [ eq : second_int_cond ] ) .
conversely let @xmath51\rightarrow\cd$ ] be lipschitz continuous as a map with values in @xmath52 and assume that @xmath53 satisfies the integral conditions ( [ eq : first_int_cond ] ) , ( [ eq : second_int_cond ] )
. then @xmath54 coincides with a trajectory of the semigroup @xmath55 .
the proof of this theorem is postponed to sections [ sec : h - riemann_solver ] and [ sec : uniqueness ] , where existence and uniqueness are proved . before these technical details
, we state the application of the above result to gas flow in section [ application ] . here
we apply theorem [ th : introduction ] to establish the existence and uniqueness of the semigroup related to pipes with discontinuous cross sections .
furthermore , we show that our approach yields the same semigroup as the approach followed in @xcite in the special case of two connected pipes .
the technical details of section [ application ] can be found at the end of the paper in section [ section5 ] .
theorem [ th : introduction ] provides an existence and uniqueness result for pipes with lipschitz continuous cross section where the equations governing the gas flow are given by @xmath56 here , as usual , @xmath57 denotes the mass density , @xmath58 the linear momentum , @xmath59 is the energy density , @xmath9 is the area of the cross section of the pipe and @xmath60 is the pressure which is related to the conserved quantities @xmath61 by the equations of state . in most situations ,
when two pipes of different size have to be connected , the length @xmath62 of the adaptor is small compared to the length of the pipes .
therefore it is convenient to model these connections as pipes with a jump in the cross sectional area .
these discontinuous cross sections however do not fulfill the requirements of theorem [ th : introduction ] .
nevertheless , we can use this theorem to derive the existence of solutions to the discontinuous problem by a limit procedure . to this end , we approximate the discontinuous function @xmath63 by a sequence @xmath64 with the following properties @xmath65\\ a^+ , & x>\frac{l}2 \end{array}\right.\ ] ] where @xmath66 is any smooth monotone function which connects the two strictly positive constants @xmath67 , @xmath68 .
one possible choice of the approximations @xmath69 as well as the discontinuous pipe with cross section @xmath9 are shown in figure [ fig:1 ] . with the help of theorem [
th : introduction ] and the techniques used in its proof , we are now able to derive the following theorem ( see also @xcite for a similar result obtained with different methods ) .
[ theorem2 ] if @xmath70 is sufficiently small , the semigroups @xmath71 related with the smooth section @xmath69 converge to a unique semigroup @xmath55 .
the limit semigroup satisfies and is uniquely identified by the integral estimates ( [ eq : first_int_cond ] ) , ( [ eq : second_int_cond ] ) with @xmath72 substituted by @xmath73 ( see section [ section5 ] ) for the point @xmath74 . more precisely let @xmath51\rightarrow\cd$ ] be lipschitz continuous as a map with values in @xmath52 and assume that @xmath53 satisfies the integral conditions ( [ eq : first_int_cond ] ) , ( [ eq : second_int_cond ] ) with @xmath72 substituted by @xmath73 for the point @xmath74
. then @xmath54 coincides with a trajectory of the semigroup @xmath55 .
observe that the same theorem holds for the @xmath75 isentropic system ( see section [ section5 ] ) @xmath76 in @xcite @xmath75 homogeneous conservation laws at a junction are considered for given admissible junction conditions .
the situation of a junctions with only two pipes with different cross sections can be modeled by our limit procedure or as in @xcite with a suitable junction condition .
if we define the function @xmath77 which describes the junction conditions as @xmath78 then it fulfills the determinant condition in ( * ? ? ?
* proposition 2.2 ) since it satisfies lemma [ lemma : riemann_prob0 ] .
here @xmath79 is the solution of the ordinary differential equation ( [ ode ] ) in section [ section5 ] . with these junction conditions
one can show that the semigroup obtained in @xcite satisfies the same integral estimate ( see the following proposition ) as our limit semigroup hence they coincide .
[ prop1 ] the semigroup defined in @xcite with the junction condition given by ( [ dynamics ] ) satisfies the integral estimates ( [ eq : first_int_cond ] ) , ( [ eq : second_int_cond ] ) with @xmath72 substituted by @xmath73 for the point @xmath74 .
the proof is postponed to section [ section5 ] .
note that proposition 1 justifies the coupling condition ( [ eq : psieq ] ) as well as the condition used in @xcite to study the riemann problem for the gas flow through a nozzle .
throughout the next two sections , we follow the structure of @xcite .
we recall some definitions and notations in there , and also the results which do not depend on the @xmath5 boundedness of the source term .
we will prove only the results which in @xcite do depend on the @xmath5 bound using our weaker hypotheses .
consider the stationary equations associated to ( [ integrable : eq : bal ] ) , namely the system of ordinary differential equations : f(v(x))_x = g(x , v(x ) ) .
[ eq : ode ] for any @xmath80 , @xmath81 , consider the initial data v(x_o ) = v. [ eq : init - data ] as in @xcite , we introduce a suitable approximation of the solutions to ( [ eq : ode ] ) , ( [ eq : init - data ] ) .
thanks to ( [ hyp : non_resonance ] ) , the map @xmath82 is invertible inside some neighborhood of the origin ; in this neighborhood , for small @xmath83 , we can define _
h(x_o , u)= f^-1 .
[ def : phi ] this map gives an approximation of the flow of ( [ eq : ode ] ) in the sense that f(_h(x_o , u))-f(u)=_0^hg(x_o+s , u)ds .
[ approx_ode ] throughout the paper we will use the landau notation @xmath84 to indicate any function whose absolute value remains uniformly bounded , the bound depending only on @xmath85 and @xmath86 .
[ lemma0 ] the function @xmath87 defined in ( [ def : phi ] ) satisfies the following uniform ( with respect to @xmath88 and to @xmath27 in a suitable neighborhood of the origin ) estimates . @xmath89
the lipschitz continuity of @xmath90 and ( [ epsilonh ] ) imply @xmath91 next we compute @xmath92\\ & \qquad\cdot \left(df(u)+\int_0^{h}d_ug(x_o+s , u)\;ds\right ) \end{split}\ ] ] which together to the identity
@xmath93 implies @xmath94-df^{-1}\left(f(u)\right)\right\|\\ & \qquad\qquad\cdot\left(\|df(u)\|+\int_0^{h}\|d_ug(x_o+s , u)\|\;ds\right)\\ & \qquad + \left\|df^{-1}\left(f(u)\right)\right\|\cdot\int_0^{h}\|d_ug(x_o+s , u)\|\;ds\\ & \le \o(1)\tilde\varepsilon_h\xrightarrow{h\to 0}0 .
\end{split}\ ] ] finally , denoting with @xmath95 the partial derivative with respect to the @xmath96 component of the state vector and by @xmath97 the @xmath98 component of the vector @xmath99 , we derive @xmath100 so that @xmath101 for any @xmath80 we consider the system ( [ integrable : eq : bal ] ) , endowed with a riemann initial datum : u(0,x)= \ { ll u _ & if x < x_o + u_r & if x > x_o . .
[ eq : riemann_data ] if the two states @xmath102 , @xmath103 are sufficiently close , let @xmath77 be the unique entropic homogeneous riemann solver given by the map u_r = ( )(u _ ) = _ n(_n) _1(_1)(u _ ) , where @xmath104 denotes the ( signed ) wave strengths vector in @xmath105 , @xcite . here
@xmath106 , @xmath107 is the shock rarefaction curve of the @xmath108 family , parametrized as in @xcite and related to the homogeneous system of conservation laws u_t+f(u)_x=0 .
[ eq : homo ] observe that , due to ( [ hyp : non_resonance ] ) , all the simple waves appearing in the solution of ( [ eq : homo ] ) , ( [ eq : riemann_data ] ) propagate with _ non - zero _ speed . to take into account the effects of the source term
, we consider a stationary discontinuity across the line @xmath109 , that is , a wave whose speed is equal to 0 , the so called . now ,
given @xmath83 , we say that the particular riemann solution : u(t , x)= \ { ll u _ & if x < x_o + u_r & if x > x_o . .
t0 [ zero - wave ] is admissible if and only if @xmath110 , where @xmath99 is the map defined in ( [ def : phi ] ) .
roughly speaking , we require @xmath102 , @xmath103 to be ( approximately ) connected by a solution of the stationary equations ( [ eq : ode ] ) .
[ def : h_r_solver ] given @xmath83 suitably small , @xmath80 , we say that @xmath53 is a @xmath111riemann solver for ( [ integrable : eq : bal ] ) , ( [ hyp : non_resonance ] ) , ( [ eq : riemann_data ] ) , if the following conditions hold * there exist two states @xmath112 , @xmath113 which satisfy @xmath114 ; * on the set @xmath115 , @xmath53 coincides with the solution to the homogeneous riemann problem ( [ eq : homo ] ) with initial values @xmath102 , @xmath112 and , on the set @xmath116 , with the solution to the homogeneous riemann problem with initial values @xmath113 , @xmath103 ; * the riemann problem between @xmath102 and @xmath112 is solved only by waves with negative speed ( i.e. of the families @xmath117 ) ; * the riemann problem between @xmath113 and @xmath103 is solved only by waves with positive speed ( i.e. of the families @xmath118 ) .
[ initial_estimates ] let @xmath80 and @xmath119 be three states in a suitable neighborhood of the origin . for @xmath111 suitably small , one has [ eq : useful_est ] |_h(x_o , u)-u| & = & (1)_0^h(x_o+s)ds , + [ eq : useful_est2 ] [ lemma : riemann_prob0 ] for any @xmath120 there exist @xmath121 , depending only on @xmath122 and the homogeneous system ( [ eq : homo ] ) , such that the following holds . for all maps @xmath123 satisfying @xmath124 and for all
@xmath125 , @xmath126 there exist @xmath127 states @xmath128 and @xmath6 wave sizes @xmath129 , depending smoothly on @xmath130 , such that with previous notations : a. @xmath131 ; b. @xmath132 ; c. @xmath133 ; d. @xmath134 .
the next lemma establishes existence and uniqueness for the @xmath111riemann solvers ( see fig.[hrs ] ) .
[ lemma : riemann_prob ] there exist @xmath135 such that the following holds : for any @xmath136 , @xmath137 $ ] , @xmath102 , @xmath138 , there exists a unique @xmath111riemann solver in the sense of definition [ def : h_r_solver ] . by lemma [ lemma0 ]
if @xmath139 is chosen sufficiently small then for any @xmath140 $ ] , @xmath80 the map @xmath141 meets the hypotheses of lemma [ lemma : riemann_prob0 ] .
finally taking @xmath142 eventually smaller we can obtain that there exists @xmath143 such that @xmath144 , for any @xmath137 $ ] . in the sequel
, @xmath145 stands for the implicit function given by lemmas [ lemma : riemann_prob0 ] and [ lemma : riemann_prob ] : @xmath146 which plays the role of a wave size vector .
we recall that , by lemma [ lemma : riemann_prob0 ] , @xmath145 is a @xmath26 function with respect to the variables @xmath147 and its @xmath26 norm is bounded by a constant independent of @xmath111 and @xmath148 .
in contrast with the homogeneous case , the wave size @xmath149 in the @xmath111riemann solver is not equivalent to the jump size @xmath150 ; an additional term appears coming from the dirac source term " ( see the special case @xmath151 ) .
[ lemma : equiv_tot_var ] let @xmath152 be the constants in lemma [ lemma : riemann_prob ] .
for @xmath153 , @xmath137 $ ] , set @xmath154 $ ] . then it holds : rcl |u_-u_r|&=&(1)(||+ ) , + [ eq : equiv_tot_var ] note that as shown in @xcite we can identify the sizes of the zero waves with the quantity = _ 0^h ( jh+s)ds .
[ size_0_wave ] with this definition all the glimm interaction estimates continue to hold with constants that depend only on @xmath85 and on @xmath86 , therefore all the wave front tracking algorithm can be carried out obtaining the existence of @xmath156-approximate solutions as defined below .
[ decadix ] given @xmath157 , we say that a continuous map @xmath158 is an @xmath159approximate solution of ( [ integrable : eq : bal])([init : data ] ) if the following holds : * as a function of two variables , @xmath160 is piecewise constant with discontinuities occurring along finitely many straight lines in the @xmath161 plane .
only finitely many wave - front interactions occur , each involving exactly two wave - fronts , and jumps can be of four types : shocks ( or contact discontinuities ) , rarefaction waves , non - physical waves and zero - waves : @xmath162 .
* along each shock ( or contact discontinuity ) @xmath163 , @xmath164 , the values of @xmath165 and @xmath166 are related by @xmath167 for some @xmath168 and some wave - strength @xmath169 .
if the @xmath170 family is genuinely nonlinear , then the lax entropy admissibility condition @xmath171 also holds .
moreover , one has @xmath172 where @xmath173 is the speed of the shock front ( or contact discontinuity ) prescribed by the classical rankine - hugoniot conditions . *
along each rarefaction front @xmath174 , @xmath175 , one has @xmath167 , @xmath176 for some genuinely nonlinear family @xmath177 .
moreover , we have : @xmath178 . *
all non - physical fronts @xmath179 , @xmath180 travel at the same speed @xmath181 .
their total strength remains uniformly small , namely : @xmath182 * the zero - waves are located at every point @xmath183 , @xmath184 .
+ along a zero - wave located at @xmath185 , @xmath186 , the values @xmath165 and @xmath166 satisfy @xmath187 for all @xmath188 except at the interaction points . * the total variation in space @xmath189
is uniformly bounded for all @xmath190 . the total variation in time
@xmath191 is uniformly bounded for @xmath192 , @xmath193 .
finally , we require that @xmath194 .
keeping @xmath83 fixed , we are about to let first @xmath195 tend to zero .
hence we shall drop the superscript @xmath111 for notational clarity .
[ first - convergence ] let @xmath196 be a family of @xmath159approximate solutions of ( [ integrable : eq : bal])([init : data ] ) .
there exists a subsequence @xmath197 converging as @xmath198 in @xmath199 to a function @xmath27 which satisfies for any @xmath200 : & & _
0^_dxdt + & & + _ 0^_j(t ,
jh)(_0^h gds)dt = 0.[eq : hequation ] moreover @xmath201 is uniformly bounded and @xmath27 satisfies the lipschitz property @xmath202 now we are in position to prove ( * ? ? ?
* theorem 4 ) with our weaker hypotheses . as in @xcite
we can apply helly s compactness theorem to get a subsequence @xmath203 converging to some function @xmath27 in @xmath204 whose total variation in space is uniformly bounded for all @xmath190 .
moreover , working as in ( * ? ? ?
* proposition 5.1 ) , one can prove that @xmath205 converges in @xmath11 to @xmath54 , for all @xmath190 .
[ th : limit_in_h ] let @xmath203 be a subsequence of solutions of equation ( [ eq : hequation ] ) with uniformly bounded total variation converging as @xmath198 in @xmath11 to some function @xmath27 . then @xmath27 is a weak solution to the cauchy problem ( [ integrable : eq : bal])([init : data ] ) .
we omit the proofs of theorem [ first - convergence ] and [ th : limit_in_h ] since they are very similar to the proofs of ( * ? ? ?
* theorem 3 and 4 ) .
we only observe that , in those proofs , the computations which rely on the @xmath5 bound on the source term have to be substituted by the following estimates . * concerning the proof of theorem [ first - convergence ] :
@xmath206 * concerning the proof of theorem [ th : limit_in_h ] : @xmath207 and @xmath208\right\}\\ & \qquad\quad+\int_{jh}^{(j+1)h}\omega(x)\left|u^h(t , x)-u(t , x)\right| .
\end{split}\ ] ] we observe that all the computations done in ( * ? ? ?
* section 4 ) rely on the source @xmath22 only through the amplitude of the zero waves and on the interaction estimates .
therefore the following two theorems still hold in the more general setting .
[ tsoin - tsoin ] there exists @xmath209 such that if @xmath210 is sufficiently small , then for any ( small ) @xmath83 there exist a non empty closed domain @xmath211 and a unique uniformly lipschitz semigroup @xmath212 whose
trajectories @xmath213 solve ( [ eq : hequation ] ) and are obtained as limit of any sequence of @xmath159approximate solutions as @xmath195 tends to zero with fixed @xmath111 .
in particular the semigroups @xmath214 satisfy for any @xmath215 , @xmath47 p^h_0 u_o = u_o , p^h_t p^h_su_o= p^h_s+tu_o , [ eq : semi_prop_ph ] p^h_t u_o - p^h_s v_o _ 1 ( ) l[eq : lip_prop_ph ] for some @xmath42 , independent on @xmath111 .
[ th : final_existence ] if @xmath210 is sufficiently small , there exist a constant @xmath42 , a non empty closed domain @xmath43 of integrable functions with small total variation and a semigroup @xmath216 with the following properties a. @xmath217 b. @xmath218 c. for all @xmath49 , the function @xmath219 is a weak entropy solution of system ( [ integrable : eq : bal ] ) .
d. for some @xmath209 and all @xmath83 small enough @xmath220 .
e. there exists a sequence of semigroups @xmath221 such that @xmath222 converges in @xmath11 to @xmath223 as @xmath224 for any @xmath225 .
[ newrem ] looking at ( * ? ? ? * ( 4.6 ) ) and the proof of ( * ? ? ?
* theorem 7 ) one realizes that the invariant domains @xmath226 and @xmath227 depend on the particular source term @xmath228 .
on the other hand estimate ( * ? ? ?
* ( 4.4 ) ) shows that all these domains contain all integrable functions with sufficiently small total variation . since the bounds @xmath84 in lemma [ lemma : equiv_tot_var ] depend only on @xmath85 and on @xmath86 , also the constant @xmath229 in (
* ( 4.4 ) ) depends only on @xmath85 and on @xmath86 .
therefore there exists @xmath230 depending only on @xmath85 and on @xmath86 such that @xmath226 and @xmath227 contain all integrable functions @xmath231 with @xmath232 .
the proof of uniqueness in @xcite strongly depends on the boundedness of the source , therefore we have to consider it in a more careful way . as in @xcite
we shall make use of the following technical lemmas whose proofs can be found in @xcite .
[ lip_on_intervals ] let @xmath233 a ( possibly unbounded ) open interval , and let @xmath234 be an upper bound for all wave speeds . if @xmath235 then for all @xmath190 and @xmath83 , one has [ eq : lip_on_intervals ] _ a+t^b - t |(p^h_t|u)(x)-(p^h_t|v)(x)|dx l_a^b ||u(x)-|v(x)|dx .
[ distance_on_intervals ] given any interval @xmath236 $ ] , define the interval of determinacy [ eq : int_of_det ] i_t=[a+t , b - t],t < .
for every lipschitz continuous map @xmath237\mapsto\cd_h(\d)$ ] and @xmath238 : [ eq : distance_on_intervals ] & & w(t)-p^h_tw(0)_1(i_t ) + & & l_0^t \{_0 } ds .
[ rem : salvezza ] lemmas [ lip_on_intervals ] , [ distance_on_intervals ] hold also substituting @xmath214 with the operator @xmath55 . in this case
we have obviously to substitute the domains @xmath211 with the domain @xmath43 of theorem [ th : final_existence ] .
let now @xmath147 be two nearby states and @xmath239 ; we consider the function [ eq : def_di_v ] v(t , x)=\ { lll u_&&x < t+x_o + u_r&&xt+x_o . .
[ lemma_stime_integrali_omogenee ] call @xmath240 the self - similar solution given by the standard homogeneous riemann solver with the riemann data ( [ eq : riemann_data ] )
. 1 . in the general case ,
one has [ eq : stima_senza_ipotesi ] 1 t_-^+ 2 . assuming the additional relations @xmath241 and @xmath242 for some @xmath243
, @xmath244 one has the sharper estimate [ eq : stima_rarefazioni ] 1 t_-^+ |v(t , x)-w(t , x)|dx = (1)^2 ; 3 .
let @xmath245 and call @xmath246 the eigenvalues of the matrix @xmath247 . if for some @xmath96 it holds @xmath248 and @xmath249 in ( [ eq : def_di_v ] ) then one has [ item : stima_per_lin ] 1 t_-^+ |v(t , x)-w(t , x)|dx = (1)|u_-u_r| ( we now prove the next result which is directly related to our @xmath111-riemann solver . [ lemma_stime_integrali_sorgente ] call @xmath240 the self - similar solution given by the @xmath111riemann solver in @xmath148 with the riemann data ( [ eq : riemann_data ] ) . 1 .
[ item : hstima_senza_ipotesi ] in the general case one has [ eq : hstima_senza_ipotesi ] 1 t_-^+ 2 .
[ item : hstima_per_lin ] assuming the additional relation @xmath250^{-1}(u^*)\displaystyle{\int_{0}^{h}}g(x_o+s , u^*)ds\ ] ] + with @xmath251 in ( [ eq : def_di_v ] ) one has the sharper estimate & & 1 t_-^+ |v(t , x)-w(t , x)|dx + & = & (1 ) ( _ 0^h(x_o+s)ds+|u_- u^*| ) _ 0^h(x_o+s)ds .
[ eq : hstima_per_lin ] _ proof .
_ estimate _ ( [ item : hstima_senza_ipotesi ] ) _ is a direct consequence of lemma [ lemma : equiv_tot_var ] .
let us prove now _ ( [ item : hstima_per_lin])_. since @xmath252 we derive & & 1 t_-^+ |v(t , x)-w(t , x)|dx + & = & 1 t_-t^0| u_-w(t , x)|dx + 1 t_0^t| u_r - w(t , x)|dx + & = & (1)=(1)|| .
this leads to ||&=&|e- e| + & = & (1)|u_r-_h(x_o , u_)| . to estimate this last term
, we define @xmath253 and compute for some @xmath254 : |u_+^-1(u^*)y_1-b(y_2,u _ ) | & & (1)|y_1||u^*-u_|+(1)|y_1-y_2| + & & + |u_+^-1(u_)y_2 -b(y_2,u_)|
. the function @xmath255^{-1}(u_\ell)y_2 -b\left(y_2,u_\ell\right)$ ] satisfies @xmath256 , @xmath257 , hence we have the estimate substitute @xmath258 then , we get @xmath259 which proves ( [ eq : hstima_per_lin ] ) . in this section
we are about to give necessary and sufficient conditions for a function @xmath260 to coincide with a semigroup s trajectory . to this end
, we prove the uniqueness of the semigroup @xmath55 and the convergence of all the sequence of semigroups @xmath214 towards @xmath55 as @xmath261 .
we begin by introducing some notations : given a @xmath262 function @xmath263 and a point @xmath264 , we denote by @xmath265 the solution of the homogeneous riemann problem ( [ eq : riemann_data ] ) with data u_=_x - u(x),u_r=_x+u(x),x_o=. [ eq : def_riemann_data ] moreover we define @xmath266 as the solution of the linear hyperbolic cauchy problem with constant coefficients w_t+aw_x = g(x ) , w(0,x)=u(x ) , [ eq : def_linear_problem ] with @xmath267 , @xmath268 .
we will need also the following approximations of @xmath266 .
let @xmath269 be a piecewise constant function .
we will call @xmath270 the solution of the following cauchy problem : @xmath271 define @xmath272 and let @xmath273 , @xmath274 , @xmath275 be respectively the @xmath276 eigenvalue , the @xmath276 right / left eigenvectors of the matrix @xmath277 . as in @xcite @xmath278 and @xmath270 have the following explicit representation @xmath279 where the function @xmath280 is defined by [ eq : definition_gh ] g_i^h(t , x)= \ { lll & _ i>0 + & _ i<0 . . using ( [ epsilonh ] )
we can compute |g_i^h(t , x)-_x-_it^x l_i , g(x)dx| = (1)_h .
hence , for any @xmath281 with @xmath282 , we have the error estimate @xmath283.\ ] ] from ( [ eq : happrox_lin_sol ] ) , ( [ eq : definition_gh ] ) , it is easy to see that @xmath284 is piecewise constant with discontinuities occurring along finitely many lines on compact sets in the @xmath285 plane for @xmath190 .
only finitely many wave front interactions occur in a compact set , and jumps can be of two types : contact discontinuities or zero waves .
the zero waves are located at the points @xmath286 , @xmath287 and satisfy w^h(t , jh+)-w^h(t , jh-)=^-1(u^ * ) _ jh^(j+1)hg(jh+s)ds .
[ eq : abra ] conversely a contact discontinuity of the @xmath276 family located at the point @xmath288 satisfies @xmath289 and
w^h(t , x_(t)+)-w^h(t , x_(t)-)=r_i(u^ * ) [ eq : cadabra ] for some @xmath290 .
now , we can state the uniqeness result in our more general setting .
[ th : characterisation ] let @xmath291 be the semigroup of theorem [ th : final_existence ] and let @xmath234 be an upper bound for all wave speeds .
then every trajectory @xmath292 , @xmath293 , satisfies the following conditions at every @xmath294 . 1 . [ first_int_cond ] for every @xmath295 , one has [ eq : first_int_cond ] _ 0 1 _
-^+ |u(+,x)-u^_(u();)(,x)|dx = 0 . 2 . [ second_int_cond ]
there exists a constant @xmath296 such that , for every @xmath297 and @xmath298 , one has [ eq : second_int_cond ] 1 _
a+^b- |u(+,x)-u^_(u ( ) ; ) ( , x)| dx + c^2 .
viceversa let @xmath51\rightarrow\cd$ ] be lipschitz continuous as a map with values in @xmath52 and assume that the conditions ( [ first_int_cond ] ) , ( [ second_int_cond ] ) hold at almost every time @xmath299
. then @xmath54 coincides with a trajectory of the semigroup @xmath55 .
[ rem : recall ] the difference with respect to the result in @xcite is the presence of the integral in the right hand side of formula ( [ eq : second_int_cond ] ) .
if @xmath35 is in @xmath5 , the integral can be bounded by @xmath300 and we recover the estimates in @xcite .
note also that the quantity @xmath301 is a uniformly bounded finite measure and this is what is needed for proving the sufficiency part of the above theorem .
* part 1 : necessity * given a semigroup trajectory @xmath302 , @xmath303 we now show that the conditions _ ( [ first_int_cond ] ) _ , _ ( [ second_int_cond ] ) _ hold for every @xmath304 . as in @xcite
we use the following notations . for fixed @xmath305
we define @xmath306 with j_t^-&=&(-(2-t+),-(t- ) ) ; + [ intervals ] j_t^o&= & ; + j_t^+&=&(+(t-),+(2-t+ ) ) .
let @xmath307 be the piecewise constant function obtained from @xmath308 dividing the centered rarefaction waves in equal parts and replacing them by rarefaction fans containing wave fronts whose strength is less than @xmath309 .
observe that : 1 t_-^+|u^,_(u();)(,x ) -u^_(u();)(,x)|dx = (1 ) .
[ eq : compare_to_eps ] applying estimate ( [ eq : distance_on_intervals ] ) to the function @xmath310 we obtain [ eq : stimapiu ] & & _ j_+|
u^,_(u();)(,x)-(p^h_u^,_(u();)(0))(x ) |dx + & & l _ ^+_0 dt .
the discontinuities of @xmath310 do not cross the dirac comb for almost all times @xmath311 .
therefore we compute for such a time @xmath312 : [ eq : uno_asterisco ] & & 1 _ j_t+ & = & 1 _
j^-_t+j^o_t+j^+_t+ |u^,_(u();)(t-+,x)-(p_^hu^,_(u ( ) ; ) ( t- ) ) ( x)|dx .
define @xmath313 the set of points in which @xmath314 has a discontinuity while @xmath315 is the set of points in which the zero waves are located .
if @xmath316 is sufficiently small , the solutions of the riemann problems arising at the discontinuities of @xmath314 do not interact , therefore @xmath317 note that the shock are solved exactly both in @xmath310 and in @xmath318 therefore they make no contribution in the summation . to estimate the approximate rarefactions we use the estimate ( [ eq : stima_rarefazioni ] ) hence @xmath319 concerning the zero waves , recall that @xmath312 is chosen such that @xmath310 is constant there , and @xmath214 is the exact solution of an @xmath111riemann problem , hence we can apply ( [ eq : hstima_senza_ipotesi ] ) with @xmath151 and obtain @xmath320 finally using ( [ eq:51 ] ) and ( [ eq:49 ] ) we get in the end [ eq : due_asterisco ] 1 _
j^o_t+ |u^,_(u();)(t-+,x)-(p_^hu^,_(u ( ) ; ) ( t- ) ) ( x)|dx + = (1)\{_j^0_t(x)dx+_h+}. moreover , following the same steps as before and using ( [ eq : stima_senza_ipotesi ] ) and ( [ eq : hstima_senza_ipotesi ] ) with @xmath151 we get [ eq : tre_asterisco ] 1 _
j^+_t+ = (1 ) \{_j^+_t(x)dx+_h } .
note that here there is no total variation of @xmath310 since in @xmath321 it is constant .
a similar estimate holds for the interval @xmath322 .
putting together ( [ eq : uno_asterisco ] ) , ( [ eq : due_asterisco ] ) , ( [ eq : tre_asterisco ] ) , one has 1 _ j_t+ |u^,_(u();)(t-+,x)-(p_^h u^,_(u();)(t- ) ) ( x)|dx + = (1 ) ( _ j_(x)dx+_h+ ) .
hence , setting @xmath323 by ( [ eq : stimapiu ] ) , we have [ eq : stimameno ] _ j_+| u^,_(u();)(,x)-(p^h_v)(x ) |dx = (1 ) ( _ j_(x)dx+_h+ ) .
finally we take the sequence @xmath221 converging to @xmath55 . using ( [ eq : lip_on_intervals ] )
we have [ after51 ] p^h_i_u()-p^h_i_v _ 1(j_+ ) & & lu()-v _ 1(j _ ) + & = & _ -2^+2|u(,x)-v(x)|dx + & = & | _ , where @xmath324 tends to zero as @xmath325 tends to zero due to the fact that @xmath326 has right and left limit at any point : for any given @xmath327 if @xmath325 is sufficiently small @xmath328 for @xmath329 . therefore by ( [ eq : compare_to_eps ] ) , ( [ eq : stimameno ] ) , we derive : & & 1 _ -^+ |u(+,x)-u^_(u();)(,x)|dx + & = & + |_+(1 ) .
the left hand side of the previous estimate does not depend on @xmath309 and @xmath330 , hence 1 _
-^+ @xmath331 depend on @xmath325 ( see [ intervals ] ) . so taking the limit as @xmath332 in the previous estimate yields ( [ eq : first_int_cond ] ) . to prove _ ( [ second_int_cond ] ) _
let @xmath333 and a point @xmath334 be given together with an open interval @xmath233 containing @xmath295 .
fix @xmath335 and choose a piecewise constant function @xmath336 satisfying @xmath337 together with @xmath338 let now @xmath270 be defined by ( [ eq : happrox_lin_sol ] ) ( @xmath339 ) .
from ( [ eq : distance_between_w_wh ] ) , ( [ eq : second_totvar_prop ] ) we have the estimate [ eq : propp ] _ a+^b- |u^_(u();)(,x)- w^h(,x ) |dx(1)(+_h(b - a ) ) . using ( [ eq : int_of_det ] ) , ( [ eq : distance_on_intervals ] ) we get [ eq : prip ] & & _
a+^b- |w^h(,x)-(p_^h w^h(0))(x ) |dx + & & l _ ^+_0 dt where we have defined @xmath340 .
let @xmath341 be a time for which there is no interaction in @xmath270 ; in particular , discontinuities which travel with a non - zero velocity do not cross the dirac comb ( this happens for almost all @xmath312 ) .
we observe that by the explicit formula ( [ eq : happrox_lin_sol ] ) : @xmath342 @xmath343 as before for @xmath316 sufficiently small we can split homogeneous and zero waves @xmath344 the homogeneous waves in @xmath270 satisfy ( [ eq : cadabra ] ) , with @xmath345 in place of @xmath346 , hence we can apply ( [ item : stima_per_lin ] ) which together with ( [ eq:3;35 ] ) , ( [ eq:3;36 ] ) leads to @xmath347 where @xmath348 denotes the jump of @xmath349 at @xmath24 .
the zero waves in @xmath270 satisfy ( [ eq : abra ] ) , hence we can apply ( [ eq : hstima_per_lin ] ) which together with ( [ eq:3;36 ] ) leads to @xmath350 let now @xmath221 be the subsequence converging to @xmath55 . since @xmath351 using ( [ eq : propp ] ) , ( [ eq : prip ] ) , ( [ eq : second_totvar_prop ] ) , and the last estimates we get & & 1 _ a+^b- |u(+,x)-u^_(u();)(,x)| dx + & & + l + & & + (1 ) \{+ ( \{|v ; ( a , b ) } + _ a^b(x)dx+_h_i)^2}. so for @xmath352 we obtain the desired inequality . + * part 2 : sufficiency * by remark [ rem : salvezza ] we can apply ( [ eq : distance_on_intervals ] ) to @xmath55 and hence the proof for the homogeneous case presented in @xcite , which relies on the property recalled in remark [ rem : recall ] , can be followed exactly for our case , hence it will be not repeated here .
@xmath353 [ [ proof - of - theorem - thintroduction ] ] proof of theorem [ th : introduction ] + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + it is now a direct consequence of theorems [ th : final_existence ] and [ th : characterisation ] .
consider the equation @xmath354 for some @xmath355 .
equation ( [ dynamics ] ) is comprised in this setting with the substitution @xmath356 . for this kind of equations we consider the exact stationary solutions instead of approximated ones as in ( [ def : phi ] ) .
therfore call @xmath357 the solution of the following cauchy problem : @xmath358^{-1}g(u(a))\\ u(0)=\bar u \end{array } \right.\ ] ] if @xmath9 is sufficiently small , the map @xmath359 satisfies lemma [ lemma : riemann_prob0 ]
. we call @xmath9-riemann problem the cauchy problem @xmath360 its solution will be the function described in definition [ def : h_r_solver ] using the map @xmath361 instead of the @xmath99 in there . observe that if @xmath362 the @xmath9-riemann solver coincides with the usual homogeneous riemann solver .
since @xmath70 , hypothesis @xmath34 is satisfied uniformly with respect to @xmath62 , moreover the smallness of @xmath367 ensures that the @xmath11 norm of @xmath35 in @xmath40 is small .
therefore the hypotheses of theorem [ th : introduction ] are satisfied uniformly with respect to @xmath62 .
let @xmath71 be the semigroup related with the smooth section @xmath69 . by remark [ newrem ]
, if @xmath368 is sufficiently small , @xmath27 belongs to the domain of @xmath71 for every @xmath369 . since the total variation of @xmath370
is uniformly bounded for a fixed initial data @xmath27 , helly s theorem guarantees that there is a converging subsequence @xmath371 . by a diagonal argument one can show that there is a converging subsequence of semigroups converging to a limit semigroup @xmath55 defined on an invariant domain ( see ( * ? ? ? * proof of theorem 7 ) ) . to zero
. we will estimate ( [ eq : first_int_condb ] ) in several steps .
first define @xmath372 and compute @xmath373 as in ( [ after51 ] ) .
then we consider the approximating sequence @xmath374 corresponding to the source term @xmath375 and the semigroups @xmath376 which converge to @xmath374 in the sense of theorem [ th : final_existence ] .
hence we have @xmath377 for notational convenience we skip the subscript @xmath96 in @xmath378 . as in ( [ eq : compare_to_eps ] ) we approximate rarefactions in @xmath379 introducing the function @xmath380 . then we define ( see figure [ fig : p ] ) @xmath381 where @xmath382 is piecewise constant with jumps in the points @xmath286 satisfying @xmath383 . furthermore @xmath384 and @xmath385 is defined as in ( [ def : phi ] ) using the source term @xmath386 .
observe that the jump between @xmath387 and @xmath388 does not satisfy any jump condition , but as @xmath382 is an `` euler '' approximation of the ordinary differential equation @xmath389 , this jump is of order @xmath390 .
since @xmath380 and @xmath391 have uniformly bounded total variation we have the estimate @xmath392 the bound @xmath84 not depending on @xmath111 .
we apply lemma [ distance_on_intervals ] on the remaining term @xmath393 to estimate this last term we proceed as before .
observe that @xmath394 does not have zero waves outside the interval @xmath395 $ ] since outside the interval @xmath396 $ ] the function @xmath397 is identically zero . if @xmath316 is small enough , the waves in @xmath398 do not interact , therefore the computation of the @xmath11 norm in the previous integral , as before can be splitted in a summation on the points in which there are zero waves in @xmath394 or jumps in @xmath399 .
observe that the jumps of @xmath400 in the interval @xmath401 , are defined exactly as the zero waves in @xmath394 so we have no contribution to the summation from this interval . outside the interval
@xmath395 $ ] , @xmath214 coincides with the homogeneous semigroup , hence we have only the second order contribution from the approximate rarefactions in @xmath399 as in ( [ eq:49 ] ) .
furthermore we might have a zero wave in the interval @xmath402 $ ] and a discontinuity of @xmath403 in the point @xmath404 of order @xmath390 . using ( [ eq : hstima_senza_ipotesi ] ) for the zero wave and ( [ eq : stima_senza_ipotesi ] ) for the discontinuity ( since @xmath214 is equal to the homogeneous semigroup in @xmath404 )
, we get @xmath405 which completes the proof if we let first @xmath406 tend to zero , then @xmath111 tend to zero , then @xmath62 tend to zero and finally @xmath325 tend to zero . as in the previous proof
, the sufficiency part can be obtained following the proof for the homogeneous case presented in @xcite .
call @xmath407 the semigroup defined in @xcite .
the estimates for this semigroup outside the origin are equal to the ones for the standard riemann semigroup see @xcite . concerning the origin we first observe that the choice ( [ eq : psieq ] ) implies that the solution to the riemann problem in ( * ? ? ?
* proposition 2.2 ) coincides with @xmath408 .
we need to show that [ eq : first_int_condbappo ] _ 01 _ -^+ |u(+,x)-|u^_u()(,x)|dx=0 . with @xmath409 . as before ,
we first approximate @xmath408 with @xmath380 and @xmath326 with @xmath410 then we apply lemma [ distance_on_intervals ] ( which holds also for the semigroup @xmath407 ) and compute & & 1 _ -^+ |(s_|v)(x)-|u^,_u()(,x)|dx + & & l1 _ ^+_0 the discontinuities of @xmath380 are solved by @xmath411 with exact shock or rarefaction for @xmath412 and with the @xmath9riemann solver in @xmath413 therefore the only difference between @xmath414 and @xmath415 are the rarefactions solved in an approximate way in the first function and in an exact way in the second .
recalling ( [ eq : stima_rarefazioni ] ) we know that this error is of second order in the size of the rarefactions .
+ to show that ( [ eq : first_int_condbappo ] ) holds , proceed as in ( [ eq:49 ] ) .
rinaldo m. colombo and mauro garavello . on the @xmath60-system at a junction . in _ control methods in pde - dynamical systems _ ,
volume 426 of _ contemp .
_ , pages 193217 .
soc . , providence , ri , 2007 . | we consider the cauchy problem for a @xmath0 strictly hyperbolic system of balance laws @xmath1 each characteristic field being genuinely nonlinear or linearly degenerate . assuming that the @xmath2 norm of @xmath3 and @xmath4 are small enough
, we prove the existence and uniqueness of global entropy solutions of bounded total variation extending the result in @xcite to unbounded ( in @xmath5 ) sources .
furthermore , we apply this result to the fluid flow in a pipe with discontinuous cross sectional area , showing existence and uniqueness of the underlying semigroup . _
2000 mathematics subject classification : _
35l65 , 35l45 , 35l60 .
_ keywords : _ hyperbolic balance laws , unbounded sources , pipes with discontinuous cross sections . |
the structure of light nuclei near the neutron drip - line is one of the interesting topic for a good number of exotic phenomena .
nuclei in this region are quite different in collectivity and clustering features than the stable counterpart in the nuclear chart . for example , the neutron magic property is lost for n = 8 in @xmath5be @xcite and n = 20 in @xmath6 mg @xcite .
the unexpectedly large reaction cross - section for @xmath7c gives the indication of neutron halo structure @xcite .
the discovery of large collectivity of @xmath8 mg by iwasaki et al .
@xcite is another example of such exotic properties .
the deformed structures , core excitation and the location of drip - line for mg and neighboring nuclei are few of the interesting properties of investigation . in this context , the discovery of @xmath0 mg and @xmath1al , once predicted to be nuclei beyond the drip - line by various mass formulae @xcite show to need the modification of the mass models . on the other hand , the appearance of n = 16 as magic number in @xmath9o and the existence of neutron halo in @xmath10li are established observations @xcite . however , the proposed proton @xcite ( @xmath11b ) and neutron @xcite halo ( @xmath12be , @xmath13b , @xmath14ne ) in the exotic nuclei are currently under investigations .
in addition to these , the cluster structure of entire the light mass nuclei and the skin formation in neutron - drip isotopes motivate us for the study of light mass drip - line nuclei . in this paper ,
our aim is to study the neutron drip - line for ne@xmath15s isotopic chain in the frame - work of a relativistic mean field ( rmf ) and nonrelativistic skyrme hartree - fock ( shf ) formalisms and analyzed the features of large quadrupole deformation of these isotopes .
the paper is organized as follows : the rmf and shf formalisms are described briefly in section ii .
the results obtained from our calculations are discussed in section iii .
finally , summary and concluding remarks are given in section iv .
the mean field methods like shf and rmf have been widely used in the study of binding energies , root mean square radii , quadrupole deformation and other bulk properties of nuclei @xcite . in general
, one can say that although the older parametrizations of shf and rmf have some limitation to predict the experimental observables , the recent forces are good enough to reproduce the bulk properties not only near the @xmath16stability line but also far away from it . here
, we use these two successful models @xcite to learn about the properties of drip - line nuclei ne@xmath15s .
the general form of the skyrme effective interaction used in the mean - field model can be expressed as a hamiltonian density @xmath17 @xcite .
this @xmath17 is written expressed as a function of some empirical parameters given as : @xmath18 where @xmath19 is the kinetic energy term , @xmath20 the zero range , @xmath21 the density dependent and @xmath22 the effective - mass dependent terms , which are relevant for calculating the properties of nuclear matter .
these are functions of 9 parameters @xmath23 , @xmath24 ( @xmath25 ) and @xmath26 , and are given as @xmath27 , \label{eq:2 } \\
{ \mathcal h}_3&=&\frac{1}{24}t_3\rho^\eta \left[(2+x_3)\rho^2 - ( 2x_3 + 1)(\rho_p^2+\rho_n^2)\right ] , \label{eq:3 } \\ { \mathcal h}_{eff}&=&\frac{1}{8}\left[t_1(2+x_1)+t_2(2+x_2)\right]\tau \rho \nonumber \\ & & + \frac{1}{8}\left[t_2(2x_2 + 1)-t_1(2x_1 + 1)\right](\tau_p \rho_p+ \tau_n \rho_n ) .
\label{eq:4}\end{aligned}\ ] ] the kinetic energy @xmath28 , a form used in the fermi gas model for non - interacting fermions .
the other terms , representing the surface contributions of a finite nucleus with @xmath29 and @xmath30 as additional parameter , are @xmath31(\vec{\nabla}\rho)^2 \nonumber\\ & & -\frac{1}{16}\left[3t_1(x_1+\frac{1}{2})+t_2(x_2+\frac{1}{2})\right ] \nonumber\\ & & \times\left[(\vec{\nabla}\rho_n)^2+(\vec{\nabla}\rho_p)^2\right ] , % \text{and } \label{eq:5 } \\ { \mathcal h}_{s\vec{j}}&=&-\frac{1}{2}\left[{b_4}\rho\vec{\nabla}\cdot\vec{j}+{b^{\prime}_4}(\rho_n\vec{\nabla}\cdot\vec{j_n } + \rho_p\vec{\nabla}\cdot\vec{j_p})\right ] .
\label{eq:6}\end{aligned}\ ] ] here , the total nucleon number density @xmath32 , the kinetic energy density @xmath33 , and the spin - orbit density @xmath34 . the subscripts @xmath35 and @xmath36 refer to neutron and proton , respectively .
the nucleon mass is represented by @xmath37 .
the @xmath38 , @xmath39 or @xmath36 , for spin - saturated nuclei , i.e. , for nuclei with major oscillator shells completely filled or empty .
the total binding energy ( be ) of a nucleus is the integral of @xmath17 .
the relativistic mean field approach is well - known and the theory is well documented @xcite . here
we start with the relativistic lagrangian density for a nucleon - meson many - body system as : @xmath40 all the quantities have their usual meanings . from the relativistic lagrangian
, we obtain the field equations for the nucleons and mesons .
these equations are solved by expanding the upper and lower components of the dirac spinor and the boson fields in an axially deformed harmonic oscillator basis .
the set of coupled equations is solved numerically by a self - consistent iteration method .
the total energy of the system in rmf formalism is given by @xmath41 where @xmath42 is the sum of the single particle energies of the nucleons and @xmath43 , @xmath44 , @xmath45 , @xmath46 , @xmath47 , @xmath48 are the contributions of the meson fields , the coulomb field , pairing energy and the center - of - mass energy , respectively . to take care of the pairing correlation for open shell nuclei , the constant gap bcs - approach is used in our calculations .
the pairing energy expression is written as : @xmath49 ^ 2,\ ] ] with @xmath50=pairing force constant , @xmath51 and @xmath52 are the occupation probabilities@xcite . the variational approach with respect to @xmath51
gives the bcs equation @xcite @xmath53 using @xmath54 .
the occupation number is defined as : @xmath55.\ ] ] the chemical potentials @xmath56 and @xmath57 are determined by the particle numbers for neutrons and protons .
the pairing energy is computed as @xmath58 . for a particular value of @xmath59 and @xmath50 ,
the pairing energy @xmath47 diverges , if it is extended to an infinite configuration space .
in fact , in all realistic calculations with finite range forces , the contribution of states of large momenta above the fermi surface ( for a particular nucleus ) to @xmath59 decreases with energy .
we use a pairing window , where the equations are extended up to the level @xmath60 .
the factor 2 has been determined so as to reproduce the pairing correlation energy for neutrons in @xmath61sn using gogny force @xcite .
the values of @xmath62 and @xmath63 are taken from @xcite , as input in the bcs - eqaution .
we compare the results with various simple and sophisticated pairing prescriptions like bcs - delta force @xcite and bcs density dependent delta force @xcite .
these calculations have been done only for @xmath64ne and @xmath65al nucleus in both ski4 and nl3 force parameter sets .
we have given these results in table [ tab : pair ] along with experimental results like quadrupole deformation parameter @xmath66 @xcite , total binding energy ( be ) @xcite and root mean square charge radius ( @xmath67 ) @xcite .
we find that , for this lighter mass region of the periodic chart , pairing is less important for majority of the cases . with pairing ,
the deformation becomes negligible for @xmath64ne and we do not get the experimental deformation parameter in rmf calculations . with no pairing ,
we reproduce substantially the deformation parameter in rmf because the _ density of states _ near fermi surface for such light nuclei are small and not conducive to pairing @xcite . to understand the influence of pairing on the open shell nuclei , we have taken into account the experimental data , wherever available .
the shf(ski4 ) results are used as guideline in the absence of these data
. we realized after comparing the calculated @xmath66 of rmf and shf with experimental data that the quadrupole deformation of shf is closer to experiment without taking pairing correlation into account .
for example , when we use the @xmath62 and @xmath63 from the experimental binding energy of odd - even values or from the empirical formula of ref .
@xcite to calculate @xmath66 for @xmath68ne in rmf(nl3 ) , we find @xmath69 , respectively for these isotopes agreeing with the result of lalazissis et al @xcite .
these @xmath66 strongly disagree with the measured values ( @xmath70 ) @xcite .
similar effects are also seen in other considered isotopes . on the other hand , if we ignore pairing , then the calculated results are often better and these @xmath66 are quite close to the experimental data .
the influence of pairing is also visible in the total binding energy . in some of the cases , even a couple of mev difference in total binding energy
is found with and without taking pairing correlation into account in rmf formalism .
contrary to the rmf , the pairing in the shf formalism is almost insensitive to quadrupole deformation for the considered mass region .
thus , we have performed the calculations through out the paper without considering pairing into account .
.[tab : pair]calculation of binding energy ( be ) , quadrupole deformation parameter @xmath66 , root mean square of matter radius ( @xmath71 ) and charge radius ( @xmath67 ) by taking various pairing methods .
we have given these results for both ski4 and nl3 parameter sets with experimental data @xcite . [ cols="^,^,^,^,^,^",options="header " , ] the @xmath3 and @xmath4 intrinsic single - particle states for the normal and superdeformed state for @xmath6 mg and @xmath8 mg .
doublets are noticed for the sd intrinsic states only .
the @xmath72 states are denoted by green lines and the @xmath73 states are denoted by black . ]
same as fig .
[ mg ] for @xmath74al and @xmath65al . ]
we discus some clear and important characteristics of superdeformed solutions ( @xmath75 or more ) obtained in mean field models as compared to the normal solutions of smaller deformation .
since the lowering and occupation of the deformation - driving @xmath76 orbits from the shell above the usual valence space is so important in producing superdeformation we have emphasized their role in this discussion .
there is the occurrence of @xmath3 , @xmath4 orbits close together in energy ( doublets ) below and near the fermi surface of the self - consistent superdeformed solutions .
this feature also occurs broadly in nilsson orbits at asymptotically large prolate deformations ( see the nilsson diagrams in bohr and mottelson vol . ii @xcite ) . in normal deformed case ,
the deformed orbits of a major shell form a `` band''-like set of orbits , distinctly separated from the major shell above and below ( see fig .
[ al ] for @xmath65al ( @xmath77=0.09 ) and @xmath74al ( @xmath77 = 0.125 ) ) .
thus physical states obtained from such intrinsic states of low deformation will be well separated in energy from those intrinsic states where excitation occurs across a major shell ( a single nucleon excitation across a major shell means a change in parity and significant energy change for small deformation ) .
the above mentioned `` band''-like separation of orbits of major shells of unique parity is quite lost in the case of superdeformation ( see fig .
[ al ] , @xmath77=0.653 of @xmath65al and @xmath77=0.660 of @xmath74al ) .
the `` band''-like orbits now spread in energy ( both downward and upward ) and orbits of successive major shells come closer to each other in energy ; an inter - mingling of orbits of different parities ( see figs . [ mg ] , [ al ] ) .
this is a significant structural change from the case of small deformation .
this has also been seen in the case of @xmath78zr in hartree - fock study @xcite .
we would like to emphasize that in the self - consistent models ( skyrme - hf and rmf ) the deformation of the nucleus is the result of the deformation of the self - consistently occupied individual orbits : @xmath79 the occupation of the more deformation driving orbits from the shell above the valence space and the unoccupation of oblate driving orbits ( e.g. @xmath80 , @xmath81 ) contribute much to configuration mixing and the lowering of @xmath82 orbits and to generation of the quadrupole deformation .
because of coming together in energy of @xmath83 and @xmath4 orbits , it is easy to see that superdeformed intrinsic states of two different parities for a particular k quantum number can be formed which will be close to each other in energy
. this will lead to parity doublets in band structures . for the neutron - rich nuclei
being discussed here , the protons are quite well bound and possible low energy excitations will be those of neutrons near the fermi surface . @xmath84 where @xmath85 and @xmath86 are the k quantum numbers for proton and neutron configurations ( k=@xmath85 + @xmath86 ) .
we illustrate schematically possible parity doublet of configurations for neutrons in fig .
[ cartoon ] , the proton configuration @xmath87 being fixed .
we show here the last few neutron occupations of superdeformed solutions and rearrangements near the fermi surface . in fig .
[ cartoon ] , ( b ) and ( c ) are parity doublet of configurations .
@xmath88 transition between ( b ) and ( c ) configurations is of odd parity multipole nature .
thus , in summary , we find a systematic behaviour of the low @xmath2 ( particularly @xmath3 and @xmath4 ) prolate deformed orbits for the superdeformed solutions .
we notice ( from the plot of the orbits ) that there is occurrence of @xmath3 and @xmath4 orbits very close to each other in energy for the superdeformed ( sd ) shape .
such @xmath3 , @xmath4 degenerate orbits occur not only for the well - bound orbits but also for the unbound states .
for example , the doublet of neutron orbits @xmath89\frac{1}{2}^+$ ] and @xmath90\frac{1}{2}^-$ ] are @xmath91 mev apart from each other in the normal deformed prolate solutions ; but they become degenerate in the superdeformed ( sd ) solutions ( shown by * in figs .
[ mg ] and [ al ] for mg , al ) .
more such doublets are easily identified ( figs .
[ mg ] and [ al ] ) for superdeformed solutions of @xmath92 mg and @xmath93al .
in fact it is noticed that the @xmath94 states of unique parity , seen clearly well separated in energy from the usual parity orbits in the normal deformed solutions , occurs closer to them in energy for the sd states , showing a degenerate parity doublet structure .
in fact , for sd solution the @xmath3 and @xmath4 orbits are intermixed in the energy plot ; while for the normal deformation they occur in distinct groups .
this is true both in the skyrme hartree fock and the rmf calculations .
this can be seen by examining the @xmath3 and @xmath4 orbits for small and large deformations in fig .
this can lead to parity mixing and octupole deformed shapes for the sd structures @xcite .
parity doublets and octupole deformation for superdeformed solutions have been discussed for @xmath95 @xcite .
there is much interest for the experimental study of the spectra of neutron - rich nuclei in this mass region @xcite .
the highly deformed structures for the neutron - rich ne - na - mg - al nuclei are interesting and signature of such superdeformed configurations ( with parity doublet structure ) should be looked for . with parity doublet of occupied orbits @xmath96 , @xmath97 ( having m= @xmath98 1/2 and + ve and -ve parities ) and
an unoccupied orbits b possible occupation of neutrons are shown in configurations ( a ) , ( b ) and ( c ) .
the two excited configurations ( b ) and ( c ) have the same @xmath86 value and represent two excited bands of different parities ( parity doublet ) .
such situation can occur for neutron configurations in superdeformed @xmath65al and @xmath6 mg , @xmath8 mg ( figs .
[ al ] , [ mg ] ) . ]
in summary , we calculate the ground and low - lying excited state properties , like binding energy and quadrupole deformation @xmath66 using rmf(nl3 ) formalism for ne , na , mg , si , p and s isotopes , near the neutron drip - line region . in general
, we find large deformed solutions for the neutron - drip nuclei which agree well with the experimental measurement .
the calculation is also repeated in the frame - work of nonrelativistic hartree - fock formalism with skyrme interaction ski4 .
both the relativistic and non - relativistic results are comparable to each other for the considered mass region . in the present calculations a large number of low - lying intrinsic superdeformed excited states are predicted in many of the isotopes and some of them are reported . from binding energy point of view , i.e. the sudden fall in @xmath99 value , the breaking of n = 28 magic number and the likely appearance of a new magic number at n = 34 noticed in our non - relativistic calculations , in contrast with rmf finding .
this is an indication of more binding than the neighbouring isotopes .
however to confirm n = 34 as a magic /non - magic number more calculations are needed .
a deformed nucleus has a collective low lying @xmath100 state .
also , a spherical nucleus can have a fairly low lying collective @xmath100 state ( e.g. sn nuclei ) because of quadrupole collectivity . in this study
we find that , for the sd shape , the low @xmath2 orbits ( particularly @xmath94 ) become more bound and nearly degenerate with the orbits of opposite parity , i.e. they show a parity doublet structure .
closely lying parity - doublet band structures and enhanced odd parity multipole transitions are possible for the superdeformed shapes .
99 a. navin et al . , phys .
85 * , 266 ( 2000 ) ; h. iwasaki et al .
b * 491 * , 8 ( 2000 ) ; h. iwasaki et al .
b * 481 * , 7 ( 2000 ) .
t. motobayashi et al . , phys .
b * 346 * , 9 ( 1995 ) .
k. tanaka et al . , phys .
lett . * 104 * , 062701 ( 2010 ) h. iwasaki et al .
b * 522 * , 227 ( 2001 ) . t. baumann et al . ,
nature * 449 * , 1022 ( 2007 ) .
p. mller , r. j. nix , w. d. myers and w. j. swiatecki , at .
data and nucl .
data tables * 59 * , 185 ( 1995 ) .
a. ozawa , t. kobayashi , t. suzuki , k. yoshida and i. tanihata , phys .
84 * , 5493 ( 2000 ) .
t. minamisono , t. ohtsubo , i. minami , s. fukuda , a. kitagawa , m. fukuda , k. matsuta , y. nojiri , s. takeda , h. sagawa and h. kitagawa , phys .
lett . * 69 * , 2058 ( 1992 ) .
i. tanihata , j. phys .
g * 22 * , 157 ( 1996 )
. t. nakamura , j. phys .
. ser . * 312 * , 082006 ( 2011 ) ; t. nakamura et al .
103 * , 262501 ( 2009 ) .
d. vautherin and d. m. brink , phys .
c * 5 * , 626 ( 1972 ) .
reinhard , rep .
52 * , 439 ( 1989 ) .
e. chabanat , p. bonche , p. hansel , j. meyer and r. schaeffer , nucl .
a * 627 * , 710 ( 1997 ) .
m. bender , paul - henri heenen and p. -g .
reinhard , rev .
phys . * 75 * , 121 ( 2003 ) .
d. lunney , j. m. pearson and c. thibault , rev .
* 75 * , 1021 ( 2003 ) .
j. r. stone , j. c. miller , r. koncewicz , p. d. stevenson and m. r. strayer , phys .
c * 68 * , 034324 ( 2003 ) . j. r. stone and p. -g .
reinhard , prog .
. phys . * 58 * , 587 ( 2007 ) .
j. erler , p. klpfel and p. -g .
reinhard , j. phys .
* 38 * , 033101 ( 2011 ) .
t. nakatsukasa , prog .
phys . 01a207
b. d. serot and j. d. walecka , adv .
* 16 * , 1 ( 1986 ) .
y. k. gambhir , p. ring and a. thimet , ann .
( n.y . ) * 198 * , 132 ( 1990 ) .
d. vretenar , a. v. afanasjev , g. a. lalazissis and p. ring , physics reports * 409 * , 101 ( 2005 ) .
j. meng et al . , prog .
. phys . * 57 * , 470 ( 2006 ) .
t. niki , d. vretenar and p. ring , prog .
* 66 * , 519 ( 2011 ) .
n. paar , d. vretenar , e. khan and gianluca col , rep .
* 70 * , 691 ( 2007 ) .
reinhard and h. f. flocard , nucl .
a * 584 * , 467 ( 1995 ) .
e. chabanat , p. bonche , p. haensel , j. meyer and r. schaeffer , nucl .
a * 635 * , 231 ( 1998 ) .
m. a. preston and r. k. bhaduri , _ structure of nucleus , addison - wesley publishing company _ , ch .
8 , page 309 ( 1982 ) .
s. k. patra , phys .
c * 48 * , 1449 ( 1993 ) . j. decharg and d. gogny , phys .
c * 21 * , 1568 ( 1980 ) .
d. g. madland and j. r. nix , nucl .
a * 476 * , 1 ( 1981 ) .
s. j. krieger , p. bonche , h. flocard , p. quentin and m. s. weiss , nucl .
a * 517 * , 275 ( 1990 ) .
h. zhang et al . , eur .
j. a * 30 * , 519 ( 2006 ) .
s. raman , c. w. jr .
nestor and p. tikkanen , at .
data and nucl .
data tables * 78 * , 1 ( 2001 ) .
m. wang , g. audi , a. h. wapstra , f. g. kondev , m. maccormick , x. xu and b. pfeiffer , chin .
c * 36 * , 1603 ( 2012 ) .
i. angeli , k. p. marinova , at .
data and nucl .
data tables * 99 * , 69 ( 2013 ) .
g. ripka , adv .
. phys . * 1 * , 183 ( 1968 ) ; w. h. bassichis , b. giraud and g. ripka , phys .
lett . * 15 * , 980 ( 1965 ) .
a. bohr and b. r. mottelson , nuclear structure vol .
i w a benjamin , inc ( 1969 ) .
g. a. lalazissis , s. raman and p. ring , at .
data and nucl .
data tables * 71 * , 1 ( 1999 ) .
s. k. patra , m. del .
estel , m. centelles and x. vias , phys .
c * 63 * , 024311 ( 2001 ) .
s. k. patra and c. r. praharaj , phys .
c * 44 * , 2552 ( 1991 ) .
j. dobaczewski , h. flocard and j. treiner , nucl .
a * 422 * , 103 ( 1984 ) .
j. meng and p. ring , phys .
* 77 * , 3963 ( 1996 ) .
m. del estal , m. centelles , x. vias and s. k. patra , phys .
c * 63 * , 044321 ( 2001 ) .
m. stoitsov , p. ring , d. vretenar and g. a. lalazissis , phys .
c * 58 * , 2086 ( 1998 ) .
h. nakada and m. sato , nucl .
phys . a * 699 * , 511 ( 2002 ) .
shan - gui zhou , j. meng and p. ring , phys .
c * 68 * , 034323 ( 2003 ) .
shan - gui zhou , j. meng , p. ring and en - guang zhao , phys .
c * 82 * , 011301(r ) ( 2010 ) .
p. arumugam , b. k. sharma , s. k. patra and r. k. gupta , phys .
c * 71 * , 064308 ( 2005 ) .
j. meng and p. ring , phy .
lett . * 77 * , 3963 ( 1996 ) .
j. p. elliott and t. h. r. skyrme , proc .
london a * 232 * , 561 ( 1955 ) .
j. w. negele , phys .
c * 1 * , 1260 ( 1970 ) .
y. sugahara and h. toki , nucl .
a * 579 * , 557 ( 1994 ) .
g. a. lalazissis , k. knig and p. ring , phys .
c * 55 * , 540 ( 1997 ) .
b. a. brown , phys . rev .
c * 58 * , 220 ( 1998 ) .
r. c. nayak and l. satpathy , at .
data and nucl .
data tables * 98 * , 616 ( 2012 ) .
p. mller , j. r. nix and k. -l .
kratz , at .
data and nucl .
data tables * 66 * , 131 ( 1997 ) .
m. samyn , s. goriely , m. bender and j. m. pearson , phys . rev .
c * 70 * , 044309 ( 2004 ) .
j. p. maharana , y. k. gambhir , j. a. sheikh and p. ring , phys .
c * 46 * , r1163 ( 1992 ) .
s. k. patra and c. r. praharaj , phys .
c * 47 * , 2978 ( 1993 ) .
f. sarazin et al . , phys .
* 84 * , 5062 ( 2000 ) .
j. l. egido , l. m. robledo , r. r. rodriguez - guzman , phys .
lett . * 93 * , 282502 ( 2004 ) .
n. onishi and s. yoshida , nucl . phy . *
80 * , 367 ( 1966 ) .
t. k. jha , m. s. mehta , s. k. patra , b. k. raj and r. k. gupta , pramana -j . phys . * 61 * , 517 ( 2003 ) ; l. satpathy and s. k. patra , nucl .
a * 722 * , 24c ( 2003 ) ; r. k. gupta , m. balasubramaniam , sushil kumar , s. k. patra , g. mnzenberg and w. greiner , j. phys .
g * 32 * , 565 ( 2006 ) ; r. k. gupta , s. k. patra and w. greiner , mod .
a * 12 * , 1327 ( 1997 ) .
t. r. werner , j. a. sheikh , w. nazarewicz , m. r. strayer , a. s. umar and m. misu , phys . lett .
b * 335 * , 259 ( 1994 ) .
ren zhongzhou , z. y. zhub , y. h. cai and xu gongou , phys .
b * 380 * , 241 ( 1994 ) .
i. hamamoto , phys .
c * 76 * , 054319 ( 2007 ) ; _ ibid _ c * 85 * , 064329 ( 2012 ) .
r. k. gupta , s. k. patra and w. greiner , mod .
a * 12 * , 1317 ( 1997 ) .
m. m. sharma , m. a. nagarajan and p. ring , phys .
b * 312 * , 377 ( 1993 ) .
r. kanungo , i. tanihata and a. ozawa , phys .
b * 528 * , 58 ( 2002 ) .
a. bohr and b. r. mottelson , nuclear structure , vol .
ii ( , chapter 5 c. r. praharaj , _
int workshop on `` nuclear many - body approaches for the @xmath101 century '' _ , september 24 - november 30 , 2007 , institute for nuclear theory website , university of washington , seattle . c. r. praharaj , _ structure of atomic nuclei _ , ch .
4 , page 108 , edited by satpathy l , narosa publishers ( delhi ) ( 1999 ) .
d. miller et al .
c * 79 * , 054306 ( 2009 ) . | the structure of ne , na , mg , al , si , p and s nuclei near the neutron drip - line region is investigated in the frame - work of relativistic mean field theory and non - relativistic skyrme hartree - fock formalism . the recently discovered nuclei @xmath0 mg and @xmath1al , which are beyond the drip - line predicted by various mass formulae
are located within these models .
we find many largely deformed neutron - rich nuclei , whose structures are analyzed . from the structure anatomy , we find that at large deformation , low @xmath2 orbits of opposite parities ( e.g. @xmath3 and @xmath4 ) occur close to each other in energy . |
we call here a _ dynamical system _ any quadruple of the form @xmath0 , where @xmath1 is a lebesgue space ( or equivalently a standard borel space equipped with a probability measure @xmath2 ) and @xmath3 an automorphism of @xmath1 : @xmath3 is a one - to - one measurable transformation of @xmath4 satisfying , for any measurable subset @xmath5 of @xmath4 , @xmath6 throughout this text , we will often use simply the symbol @xmath3 to designate the dynamical system @xmath0 .
we will also often need a second dynamical system @xmath7 , which will stand for the quadruple @xmath8 such dynamical systems , which are the objects of interest in measurable ergodic theory , can be considered from a rather probabilistic viewpoint by studying _
stationary processes_. indeed , any measurable map @xmath9 gives rise to a stationary process @xmath10 defined by @xmath11 .
( the set @xmath12 in which @xmath13 takes its values could be any standard borel space ; most of the time , it will be a finite or countable alphabet . )
conversely , to any @xmath12-valued stationary process @xmath10 , we can associate in a canonical way a dynamical system : take @xmath14 , the sample space of the whole process , equipped with its borel @xmath15-algebra @xmath16 and the law @xmath17 of the process . since @xmath13 is stationary , @xmath18 is invariant by the shift @xmath19 : @xmath20 where @xmath21 . in the dynamical system @xmath0 , the stationary process @xmath22
is called a _ generating process _ if the @xmath15-algebra it generates is , modulo @xmath2 , the whole @xmath15-algebra @xmath23 .
observe that in the dynamical system @xmath24 , the process @xmath13 ( obtained by looking at the coordinates ) is always a generating process .
the two dynamical systems @xmath3 and @xmath7 are said to be _ isomorphic _ if we can find invariant subsets @xmath25 , @xmath26 , with @xmath27 , and a measurable one - to - one map @xmath28 with , for all @xmath29 , @xmath30 and such that @xmath31 . if @xmath13 is a generating process for the dynamical system @xmath3 , then @xmath3 is isomorphic to the dynamical system @xmath32 defined above .
recall the following well - known result of ergodic theory , which shows that studying dynamical systems is essentially the same thing as studying stationary processes taking values in a countable alphabet .
( a nice proof can be found in @xcite . )
[ generating - process - theorem ] every dynamical system admits a generating process taking its values in a countable alphabet . in his famous article initiating the theory of joinings @xcite
, furstenberg observes that a kind of arithmetic can be done with dynamical systems : indeed , there are two natural operations in ergodic theory which present some analogy with the integers . first , if we are given two dynamical systems @xmath0 and @xmath8 , we can construct their _ direct product _
@xmath33 where @xmath34 .
note that the cartesian product @xmath35 is here equipped with the direct product of the probability measures @xmath2 and @xmath36 , which is always @xmath37-invariant .
acting on equivalence classes of isomorphic dynamical systems , the direct product operation is commutative , associative , and possesses a neutral element which is the trivial system reduced to one singleton . translated in the context of stationary processes , if @xmath13 and @xmath38 are respectively @xmath12-valued and @xmath39-valued stationary processes ( not necessarily defined on the same probability space ) , their direct product @xmath40 is the @xmath41-valued stationary process whose law is the direct product of the laws of @xmath13 and @xmath38 .
that is to say , we are making the two processes @xmath13 and @xmath38 live together in an independant ( hence stationary ) way .
second , we say that @xmath8 is a _ factor _ of @xmath0 if we can find a measurable map @xmath42 satisfying * @xmath43 ; * @xmath44 .
such a map is called a _ homomorphism _ of dynamical systems .
heuristically , the existence of such a homomorphism means that we can see the system @xmath7 inside the system @xmath3 by looking at @xmath45 . to this factor
is associated the _ factor @xmath15-algebra _ @xmath46 .
this @xmath15-algebra has the property that if @xmath5 is measurable with respect to @xmath46 , then so are @xmath47 and @xmath48 .
conversely , each sub-@xmath15-algebra @xmath49 of @xmath23 which is invariant by @xmath3 and @xmath50 is a factor @xmath15-algebra .
indeed , such a @xmath15-algebra is always generated by some stationary process @xmath13 living in the system @xmath3 ( this is an adaptation of theorem [ generating - process - theorem ] to the case of a factor @xmath15-algebra ) , and the dynamical system @xmath32 associated to @xmath13 is a factor of @xmath3 : take @xmath51 .
clearly , both systems @xmath7 and @xmath3 are factors of the direct product @xmath37 : these systems are seen by looking at the two coordinates in the cartesian product .
the analogy with the integers introduced by furstenberg now gives two ways to define @xmath3 and @xmath7 are relatively prime " .
[ no - common - factor - property ] @xmath3 and @xmath7 have no common factor except the trivial system reduced to one point .
[ disjointness - property ] each time @xmath3 and @xmath7 appear as factors in some dynamical system , then their direct product @xmath37 also appears as a factor should be stated in the following , stronger form : each time @xmath3 and @xmath7 appear as factors in some dynamical system through the respective homomorphisms @xmath52 and @xmath53 , @xmath37 also appears as a factor through a homomorphism @xmath54 such that @xmath55 and @xmath56 , where @xmath57 and @xmath58 are the projections on the coordinates in the cartesian product @xmath35 . ] .
the relationships between these two properties are exposed in section [ joinings - factors - section ] .
property [ disjointness - property ] has been called by furstenberg the _ disjointness _ property of @xmath3 and @xmath7 .
the study of disjointness turns out to be quite enlightening , as it refers to _ all the possible ways that two systems can be seen living together inside a third one_. as explained in the next subsection , this is precisely the theory of joinings .
the large scale of applications of disjointness and joinings is already visible in furstenberg s article where the concepts of disjointness and joinings where introduced .
indeed , furstenberg s initial motivations ranged from the classification of dynamical systems ( how disjointness can be used to characterized some classes of dynamical systems ) to a question in diophantine approximation ( which multiplicative semigroups of integers @xmath7 have the property that any real number is the limit of a sequence of rational numbers with denominators belonging to @xmath7 ? ) , passing through a filtering problem in probability theory ( if @xmath59 and @xmath60 are two real - valued stationary processes , can we recover @xmath59 from the knowledge of @xmath61 ? ) . since furstenberg
s article , disjointness and joinings have been widely studied , and many other related notions have been introduced . in the present work ,
we shall mainly concentrate on some links between joinings and other ergodic properties of dynamical systems . for a more complete treatment of ergodic theory via joinings
, we refer the readers to eli glasner s book @xcite .
let @xmath0 and @xmath8 be two dynamical systems .
a _ joining _ of @xmath3 and @xmath7 is a probability measure @xmath62 on the cartesian product @xmath35 , whose marginals on @xmath4 and @xmath63 are @xmath2 and @xmath36 respectively , and which is invariant by the product transformation @xmath37 .
the set of joinings of @xmath3 and @xmath7 , denoted by @xmath64 , is never empty since it always contains the product measure @xmath65 .
each @xmath66 provides a new dynamical system @xmath67 , which will be denoted by @xmath68 to avoid confusion with the direct product @xmath37 .
both @xmath3 and @xmath7 are factors of @xmath68 .
conversely , if @xmath3 and @xmath7 appear as factors in a third system @xmath69 through respectively the maps @xmath70 and @xmath71 , we can consider the map @xmath72 defined by @xmath73 .
then the probability measure @xmath74 on @xmath35 satisfies the conditions to be a joining of @xmath3 and @xmath7 , and the system @xmath68 is a factor of @xmath75 .
the set of joinings of @xmath3 and @xmath7 is equipped with the topology defined by @xmath76 this topology is metrizable : an example of distance defining it is given by @xmath77 where @xmath78 and @xmath79 are countable algebras generating respectively the @xmath15-algebras @xmath23 and @xmath80 . with this topology , the set of joinings @xmath64 is turned into a _
compact metrizable topological space_. it is interesting to observe that , when @xmath4 and @xmath63 are compact metric spaces , this topology coincides with the weak topology on the set of joinings .
all these definitions can be extended in an obvious way to the notion of _ joining of a finite or countable family of dynamical systems _
@xmath81 . if all the @xmath82 s are copies of a single dynamical system @xmath3 , we speak of _ self - joinings _ of @xmath3 .
the set of self - joinings of order @xmath83 of @xmath3 ( joinings of @xmath83 copies of @xmath3 ) is denoted by @xmath84 , and we simply write @xmath85 for @xmath86 .
we denote by @xmath87 the set of _ ergodic _ joinings of @xmath3 and @xmath7 .
since every factor of an ergodic system is still ergodic , a necessary condition for @xmath87 not to be empty is that both @xmath3 and @xmath7 be ergodic .
the following proposition shows that the converse is true .
given two ergodic dynamical systems @xmath3 and @xmath7 , there exists at least one ergodic joining of @xmath3 and @xmath7 .
we start from the only joining whose existence is known , @xmath88 .
this joining may not be ergodic , but in this case we consider its _ decomposition in ergodic components _ : there exists a probability measure @xmath89 on the set of all @xmath37-invariant ergodic probability measures such that @xmath90 denoting by @xmath91 ( respectively @xmath92 ) the marginals of the ergodic component @xmath93 on @xmath4 ( respectively @xmath63 ) , we get @xmath94 and @xmath95 since @xmath2 and @xmath36 are ergodic , this implies that for @xmath89-almost every @xmath96 , @xmath97 and @xmath98 . in other words , for @xmath89-almost every @xmath96 , @xmath93 is an ergodic joining of @xmath3 and @xmath7 .
note that this proof implies a stronger result : when @xmath3 and @xmath7 are ergodic , every joining of @xmath3 and @xmath7 is a convex combination of ergodic joinings . since an ergodic measure can not be written as a non trivial convex combination of invariant measures , we get the following .
[ characterization - of - ergodic - joinings ] if @xmath3 and @xmath7 are ergodic , the set of ergodic joinings of @xmath3 and @xmath7 is the set of extremal points in the compact convex set @xmath64 .
in this section , we present the two extreme cases concerning the joinings of two dynamical systems @xmath3 and @xmath7 .
the first one is when the set of joinings @xmath64 is reduced to the singleton @xmath99 .
heuristically , this means that @xmath3 and @xmath7 have nothing in common which could lead to a non - trivial joining ( what they have in common when they are not disjoint is developped in sections [ dlcf - subsection ] and [ t - factor - subsection ] ) . to the opposite ,
if @xmath3 and @xmath7 are isomorphic , we can construct very special joinings of @xmath3 and @xmath7 , namely the joinings supported on graphs of isomorphisms .
[ disjointness - definition ] the dynamical systems @xmath0 and @xmath8 are _ disjoint _ if the product measure @xmath65 is the only joining of @xmath3 and @xmath7 .
we write in this case @xmath100 .
we leave as an exercise to the reader the verification of the equivalence of this definition with the strong form of property [ disjointness - property ] . by the way , why was it necessary to state this strong form ?
( hint : there exists a non - trivial dynamical system @xmath3 which is isomorphic to the direct product @xmath101 . )
we now give the most elementary example of disjointness : the identity is disjoint from any ergodic dynamical system .
if @xmath3 is the identity map on @xmath4 and @xmath7 is any ergodic dynamical system , then @xmath100 .
let @xmath62 be any joining of @xmath102 and @xmath7 .
for any @xmath103 , @xmath29 , invariance of @xmath62 by @xmath37 gives @xmath104 since @xmath7 is ergodic , @xmath105 converges to @xmath106 @xmath36-almost everywhere , hence @xmath62-almost everywhere
. this implies @xmath107 hence @xmath108 . another useful way to state
the preceding result is the following : [ id - times - s - invariance ] let @xmath3 and @xmath7 be two dynamical systems , with @xmath7 ergodic , and @xmath66 . if @xmath62 is invariant by @xmath109 , then @xmath108 . as a nice application of proposition
[ id - times - s - invariance ] , we give below an elegant proof of furstenberg s multiple recurrence theorem in the ( easy ) case of weakly mixing transformations @xcite .
this proof was given by v. ryzhikov in @xcite .
recall that @xmath3 is weakly mixing if @xmath101 is ergodic , and that this implies * @xmath37 is ergodic for any ergodic system @xmath7 ( see also subsection [ subsection - product - ergodicity ] ) ; * @xmath110 is weakly mixing for any @xmath111 .
let @xmath3 be a weakly mixing dynamical system .
for any integer @xmath112 , and any measurable subsets @xmath113 , we have @xmath114 consider for any @xmath115 the self - joining @xmath116 of @xmath3 of order @xmath117 defined by @xmath118 note that the result we want to prove is equivalent to the following translation in the language of joinings : @xmath119 by compacity , it is enough to verify that if @xmath62 is any cluster point of the sequence @xmath120 , then @xmath62 is the product measure , which we shall prove by induction on @xmath83 .
observe that such a cluster point is always @xmath121-invariant .
thus proposition [ id - times - s - invariance ] gives the result when @xmath122 by ergodicity of @xmath3 .
now take @xmath123 and suppose the result is true for @xmath124 .
this induction hypothesis ensures that the marginal of @xmath62 on the last @xmath83 coordinates is the product measure .
hence @xmath62 can be seen as a joining of @xmath125 with @xmath126 .
since @xmath126 is ergodic ( because @xmath3 is weakly mixing ) , proposition [ id - times - s - invariance ] shows that the result is still true for @xmath83 . as we have already mentioned , one of the reasons for the introduction of disjointness by furstenberg was the classification of classes of dynamical systems . and
indeed there are many results in this direction : for example , furstenberg proved that a dynamical system has zero entropy if and only if it is disjoint from any bernoulli dynamical system .
it also follows from his work that @xmath3 is weakly mixing if and only if it is disjoint from any rotation on the circle .
some further explanations on how disjointness can characterize such classes of dynamical systems will be given in section [ stable - subsubsection ] ( see in particular theorem [ disjoint - classes - theorem ] ) .
in @xcite was introduced the so - called _ weak disjointness _
property of dynamical systems : @xmath0 and @xmath8 are said to be _ weakly disjoint _ if , given any function @xmath127 in @xmath128 and any function @xmath129 in @xmath130 , there exist a set @xmath5 in @xmath23 and a set @xmath131 in @xmath80 such that * @xmath132 * for all @xmath133 and for all @xmath134 , the sequence @xmath135 converges .
( in fact it is sufficient to check this statement for only dense families of @xmath127 and @xmath129 in their respective @xmath136 spaces . )
the main motivation for the introduction of this property was the study of non - conventional ergodic averages : suppose that @xmath7 and @xmath3 act on the same probability space @xmath4 , and consider for @xmath127 and @xmath129 in @xmath137 the averages @xmath138 when do these averages converge @xmath2-a.e . ?
if we assume that @xmath3 and @xmath7 are weakly disjoint , then almost - everywhere pointwise convergence holds .
as we could guess from the name of the property , if @xmath3 and @xmath7 are disjoint , they are weakly disjoint .
the link between joinings and convergence of ergodic sums like in comes from the following remark : we can always assume that @xmath3 and @xmath7 are continuous transformations of compact metric spaces @xmath4 and @xmath63 respectively ( indeed , any measurable system is isomorphic to such a transformation on a compact metric space : see e.g. @xcite ) .
then the set of probability measures on @xmath35 equipped with the topology of weak convergence is metric compact .
if moreover @xmath3 and @xmath7 are ergodic , we can easily find subsets @xmath25 and @xmath26 with @xmath27 , such that for all @xmath139 , any cluster point of the sequence @xmath140 is automatically a joining of @xmath3 and @xmath7 .
when @xmath3 and @xmath7 are disjoint , there is therefore only one possible cluster point to the sequence @xmath141 which is @xmath142 .
this ensures that for continuous functions @xmath127 and @xmath129 , convergence of to the product of the integrals of @xmath127 and @xmath129 holds , hence @xmath3 and @xmath7 are weakly disjoint .
it is not difficult to show that weak disjointness is really weaker than disjointness . indeed , there are many examples of dynamical systems which are self - weakly disjoint ( weakly disjoint from themselves ) : for example , irrational rotations and chacon s transformation are self - weakly disjoint ( see @xcite for details and other examples )
. however , a non trivial dynamical system is never disjoint from itself , as we will see in the next paragraph .
suppose that @xmath7 is a factor of @xmath3 , with @xmath42 a homomorphism .
then @xmath143 gives rise to the existence of a very special joining of @xmath3 and @xmath7 , which we denote by @xmath144 , and which is defined by @xmath145 [ factor_joining_lemma ] the joining @xmath144 has the following two properties : * @xmath146 , where @xmath147 is the graph of @xmath143 ; * @xmath148 , where the notation ` @xmath149 ' means that for each @xmath150 in the @xmath15-algebra @xmath151 , there exists a @xmath152 in the @xmath15-algebra @xmath153 such that @xmath154 .
take a refining and generating sequence @xmath155 of finite measurable partitions in @xmath156 .
( that is to say , @xmath157 is finer than @xmath158 , and if @xmath159 and @xmath160 are distinct points in @xmath63 , there exists some @xmath161 for which @xmath159 and @xmath160 do not belong to the same atom of @xmath158 ; this ensures in particular that the @xmath15-algebra generated by the partitions @xmath158 is @xmath80 . )
then @xmath162 can be written as @xmath163 but for any @xmath161 , it is clear from the definition of @xmath144 that @xmath164 which proves that @xmath144 is supported on the graph of @xmath143 . for the second property ,
it is easy to verify that @xmath144 identifies any @xmath165 , where @xmath29 , with @xmath166 .
it is quite remarkable to see that the converse is true , _ i.e. _ we can characterize the fact that @xmath7 is a factor of @xmath3 by the existence of some special joining . [ factor_joining ] suppose that there exists a joining @xmath66 such that @xmath167 then @xmath7 is a factor of @xmath3 and we can find a homomorphism @xmath42 such that @xmath62 is supported on the graph of @xmath143 .
take again a refining and generating sequence @xmath155 of finite measurable partitions in @xmath156 .
for each @xmath161 , @xmath158 can be identified via @xmath62 with some @xmath168 in @xmath4 , where @xmath169 is a refining , but not necessarily generating , sequence of partitions in @xmath170 , now , for @xmath2-almost every @xmath171 , if we denote by @xmath172 the atom of @xmath168 to which @xmath173 belongs and @xmath174 the atom of @xmath175 identified via @xmath62 with @xmath172 , then @xmath176 is a singleton .
denote by @xmath45 the only element of this set ; we let the reader check that @xmath177 satisfies the required properties .
if we assume further that @xmath3 and @xmath7 are _ isomorphic _ ( in other words , the homomorphism @xmath143 from @xmath4 to @xmath63 is invertible and @xmath178 is also a homomorphism of dynamical systems ; in this case we will rather use the symbol @xmath179 instead of @xmath143 ) , the joining @xmath180 now satisfies @xmath181 , where the notation ` @xmath182 ' means that we have both @xmath149 and @xmath183 .
the analog of proposition [ factor_joining ] is also true , and we summarize all these results in the following theorem .
[ joinings - supported - on - graphs - thm ] let @xmath3 and @xmath7 be two dynamical systems .
@xmath7 is a factor of @xmath3 if and only if there exists some joining @xmath66 such that @xmath184 ; in this case @xmath62 is supported on the graph of some homomorphism . @xmath3 and @xmath7 are isomorphic if and only if there exists some joining @xmath66 such that @xmath185 ; in this case @xmath62 is supported on the graph of some isomorphism .
there are very deep applications of the characterization of isomorphic systems by joinings .
krieger s theorem , stating that any dynamical system with entropy less than @xmath186 is isomorphic to the shift map on @xmath187 with an appropriate invariant measure , and ornstein s theorem characterizing the dynamical systems which are isomorphic to some bernoulli shift , can both be proved using this result ( see @xcite , chapter 7 ) .
here we illustrate the power of theorem [ joinings - supported - on - graphs - thm ] by giving lemaczyk s elegant proof of a well - known theorem due to halmos and von neumann .
this proof is taken from @xcite . recall that a system @xmath3 is said to have _ discrete spectrum _ if the subspace of @xmath136 spanned by the eigenvectors of @xmath188 is dense in @xmath136 .
if @xmath3 is ergodic and has discrete spectrum , and if @xmath7 is spectrally isomorphic to @xmath3 , then @xmath3 and @xmath7 are isomorphic . first note that @xmath7 is ergodic and also has discrete spectrum with the same eigenvalues as @xmath3 , since it is spectrally isomorphic to @xmath3 .
let @xmath62 be an ergodic joining of @xmath3 and @xmath7 .
for an eigenvalue @xmath189 of @xmath3 , let @xmath127 be an eigenvector associated to @xmath189 in the system @xmath3 , and @xmath129 associated to @xmath189 in the system @xmath7 .
then both functions @xmath190 and @xmath191 are eigenvectors in @xmath68 .
since @xmath68 is an ergodic system , @xmath189 is a simple eigenvalue in @xmath68 , hence @xmath191 and @xmath190 are proportional .
this implies that @xmath190 belongs to @xmath192 ; but since eigenvectors are dense in @xmath193 , this in turn implies @xmath194 .
for the same reason , the inverse inclusion is also true , which proves that @xmath185 , and the two systems are isomorphic . when studying self - joinings of a single dynamical system @xmath3 , we are obviously in a situation where the two systems that we want to join together are isomorphic .
now , thinking of what an isomorphism between @xmath3 and itself is , we see that this has to be an automorphism of the lebesgue space @xmath1 , which commutes with @xmath3 .
we call _ commutant of @xmath3 _ , and denote by @xmath195 , the set @xmath196 for each @xmath197 , let us denote by @xmath198 the self - joining of @xmath3 defined by @xmath199 this leads to an identification of @xmath195 with some subset of @xmath86 , namely the subset of all @xmath200 such that @xmath201 observe that for each @xmath3 the commutant of @xmath3 contains at least the powers @xmath202 , @xmath203 .
besides , note that each joined system @xmath204 for @xmath197 is isomorphic to @xmath3 ( an isomorphism is given by @xmath205 ) . in particular ,
@xmath204 is ergodic as soon as @xmath3 is ergodic .
this gives for each ergodic @xmath3 a family of `` obvious '' ergodic self - joinings : the @xmath206 , @xmath203 . in the case where @xmath3 is weakly mixing , the product measure @xmath207 is also considered as an obvious ergodic self - joining . in 1979 , rudolph introduced in @xcite the important notion of `` minimal self joinings '' , which says that @xmath3 has no other ergodic self - joinings of order 2 than the obvious ones .
the notion of `` @xmath83-fold minimal self - joinings '' refers to joinings of order @xmath83 : for @xmath208 , we say that an ergodic @xmath3 has _
@xmath83-fold minimal self - joinings _ if for each ergodic joining @xmath62 of @xmath83 copies of @xmath3 , we can partition the set @xmath187 of coordinates into two subsets @xmath209 such that 1 .
[ first - condition ] for @xmath210 and @xmath211 belonging to the same @xmath212 , the marginal of @xmath62 on the coordinates @xmath210 and @xmath211 is some @xmath206 ; 2 . for @xmath213 ,
the coordinates @xmath214 are independent .
we say that @xmath3 has minimal self - joinings ( msj ) if @xmath3 has @xmath83-fold minimal self - joinings for every @xmath215 .
in fact , it has been proved by glasner , host and rudolph @xcite that for a weakly mixing @xmath3 , 3-fold minimal self - joinings implies @xmath83-fold minimal self - joinings for all @xmath83 .
( see also a proof in @xcite . )
a very important question which remains open in the area is the following : does there exist a transformation @xmath3 which has 2-fold but not 3-fold minimal self - joinings ? ( this question is closely related to the problem of 2-fold and 3-fold mixing : see section [ subsection-2 - 3-mixing ] . )
there are many examples of dynamical systems with the msj property , the simplest of which is probably the well known chacon s transformation ( see @xcite ) .
one immediate consequence of this property is that @xmath195 is reduced to the powers of @xmath3 .
this in turn implies that the transformation @xmath3 has no square root ( _ i.e .
there is no automorphism @xmath7 such that @xmath216 _ ) .
it can be noticed that the first example of a transformation with no square root , given by ornstein @xcite in 1970 , belongs to the class of _ mixing rank - one systems _ , which all turned out to have the msj property , as king proved in 1988 @xcite .
another joining proof of this result has been given by ryzhikov @xcite .
if , in condition ( [ first - condition ] ) of the definition of msj , we replace @xmath206 by @xmath198 , for some @xmath197 , we get the weaker , but important , notion of _ simplicity_. in the case of joinings of order 2 , this property was introduced by veech @xcite , who proved a useful result on the structure of factor @xmath15-algebras of 2-simple systems .
the general definition of simplicity was later given by del junco and rudolph @xcite .
we also refer to thouvenot @xcite for a presentation of properties of simple systems . as a nice application of joinings to the study of @xmath195 , we should also mention here ryzhikov s proof of king s so - called _ weak closure theorem _ , stating that if @xmath3 is rank one , any @xmath7 in @xmath195 is the limit of some sequence @xmath217 of the powers of @xmath3 @xcite .
in this section we discuss the links and the differences between the two properties proposed in the introduction to define `` @xmath3 and @xmath7 are relatively prime '' .
the starting point of the theory is the existence of a special joining arising when @xmath3 and @xmath7 have a common factor .
we first define this special joining in the particular case of two systems which obviously have a common factor , since they both arise as a joining of a fixed system @xmath75 with other systems .
let @xmath0 , @xmath8 and @xmath69 be three dynamical systems .
suppose that @xmath218 is a joining of @xmath3 and @xmath75 , and @xmath219 is a joining of @xmath7 and @xmath75 .
then we can put all these systems together so that the marginal on @xmath220 is @xmath218 and the marginal on @xmath221 is @xmath219 : first pick @xmath222 according to the probability law @xmath223 , then pick @xmath173 and @xmath159 according to their conditionnal law knowing @xmath222 in the respective joinings @xmath218 and @xmath219 , but independently of each other .
more precisely , consider the 3-fold joining of @xmath3 , @xmath7 and @xmath75 denoted by @xmath224 defined by setting , for all @xmath225 , @xmath29 and @xmath226 @xmath227 \ , \ee_{\lambda_s}[\ind{y\in b}|z ] \
, d\rho(z).\ ] ] we get in this way a joining of the two systems @xmath228 and @xmath229 identifying the coordinate @xmath222 on both systems .
this joining is called the _ relatively independent joining of @xmath228 and @xmath229 above their common factor @xmath75_. now turn to the more general situation of two systems @xmath3 and @xmath7 sharing a common factor @xmath75 .
this means that we have homomorphisms of dynamical systems @xmath230 and @xmath231 , which give us special joinings @xmath232 and @xmath233 ( see section [ factor_joining_subsection ] ) .
we can then consider the relatively independent joining @xmath234 , which is a 3-fold joining of @xmath3 , @xmath7 and @xmath75 .
the marginal of @xmath235 on @xmath35 is now a joining of @xmath3 and @xmath7 , called the _
relatively independent joining of @xmath3 and @xmath7 above their common factor @xmath75 _ , and denoted by @xmath236 .
this joining has the strong property of identifying the projections of both systems on the common factor . in summary
, the @xmath15-algebra @xmath151 can be viewed as a sub-@xmath15-algebra of both @xmath23 and @xmath80 , any @xmath151-measurable function can be viewed as a function defined on @xmath237 , and we have the following integration formula : @xmath238(z)\,\ee[g(y)|\c](z)\,d\rho(z).\ ] ] ( 0,0 ) and @xmath7 above their common factor @xmath75 : given a ( common ) fixed projection on the factor , the joining is the direct product of the two conditional measures.,title="fig : " ] # 1#2#3#4#5 @font ( 5402,3152)(225,-2627 ) ( 1021,-2041)(0,0)[lb ] ( 2595,-2268)(0,0)[lb ] ( 4141,-1906)(0,0)[lb ] ( 2971,-196)(0,0)[lb ] ( 2351,-308)(0,0)[lb ] ( 2656,238)(0,0)[lb ] [ rij - prop ] let @xmath239 be the relatively independent joining of @xmath3 and @xmath7 above their common factor @xmath75 .
then @xmath240 by construction , @xmath239 is the marginal on @xmath35 of the 3-fold joining @xmath235 of @xmath3 , @xmath7 and @xmath75 whose marginal on @xmath241 ( respectively @xmath242 ) is @xmath243 ( respectively @xmath244 ) .
recall that , from lemma [ factor_joining_lemma ] , we have @xmath245 @xmath243-a.e . and @xmath246 @xmath244-a.e .
therefore , @xmath247 @xmath235-a.e . , hence @xmath248 @xmath239-a.e .
note that what we did here with two systems could have been done with any finite or countable family of systems sharing a common factor .
any factor @xmath7 of @xmath3 can be considered as a common factor of two copies of @xmath3 , thus gives rise to a special self - joining of @xmath3 : the relatively independent self - joining of @xmath3 above its factor @xmath7 . as in ordinary arithmetic of integer numbers , every dynamical system which is not reduced to one point has at least two factors : @xmath3 itself is of course a factor of @xmath3 , and the system reduced to one point is also a factor of @xmath3 .
note that the relatively independent joining of @xmath3 above @xmath3 itself is the self - joining @xmath249 of @xmath3 supported on the graph of @xmath125 , and that the relatively independent joining of @xmath3 above the factor reduced to one point is the product measure @xmath207 .
now , we want to show that any other factor of @xmath3 is incompatible with the 2-fold msj property .
let @xmath0 be an ergodic dynamical system where @xmath4 is not finite .
if @xmath3 has another non - trivial factor than @xmath3 itself , then @xmath3 does not have the 2-fold msj property .
let @xmath7 be a non - trivial factor of @xmath3 , and consider @xmath250 the relatively independent self - joining of @xmath3 above @xmath7 constructed from the homomorphism @xmath42 .
the system defined by this joining may not be ergodic , but if we assume further that @xmath3 has the 2-fold msj property , the ergodic decomposition of @xmath250 has the form @xmath251 where @xmath252 , @xmath253 and @xmath254 .
denote by @xmath173 and @xmath255 the two coordinates in the cartesian square of @xmath4 , and remember that , by proposition [ rij - prop ] , we have @xmath256 @xmath250-a.e . since @xmath7 is not the factor reduced to one point , this would not be possible if @xmath207 appeared in the ergodic decomposition , hence @xmath257 .
moreover , if the ergodic decomposition of @xmath250 was reduced to @xmath249 , @xmath7 would be @xmath3 itself .
hence there is some @xmath258 such that @xmath259 .
in other words , there exists some @xmath258 such that , under @xmath250 , @xmath260 this implies that , in the factor @xmath7 , @xmath261 with positive probability . but @xmath7 is ergodic ( because @xmath3 is ) , hence the space @xmath63 on which @xmath7 acts is finite . in turn
, this implies that there exists at least one @xmath262 such that @xmath263 now , observe that @xmath250 conditioned on @xmath264 is the direct product of @xmath2 conditioned on @xmath264 with itself . thus , implies that @xmath2 conditioned on @xmath264 has at least one atom .
but @xmath265 , hence @xmath2 itself has an atom , hence @xmath3 is periodic .
if @xmath3 is ergodic aperiodic with the 2-fold msj property , then any non constant stationary process living in the system @xmath3 generates the whole @xmath15-algebra .
we now turn back to the case of two dynamical systems @xmath3 and @xmath7 sharing a common factor @xmath75 . as we have already noticed in the case of self - joinings , proposition [ rij - prop ] ensures that if @xmath75 is not reduced to one point , the relatively independent joining of @xmath3 and @xmath7 above @xmath75 is not the product measure @xmath65 , hence @xmath3 and @xmath7 are not disjoint .
this means that property [ disjointness - property ] ( the disjointness property ) implies property [ no - common - factor - property ] ( the lack of common factor ) .
it had been asked by furstenberg @xcite whether these two properties were , in fact equivalent .
the discovery by rudolph of systems with the msj property allowed him to answer negatively to this question @xcite .
there exist two non disjoint systems @xmath7 and @xmath3 without any other common factor than the trivial one .
let @xmath266 be a stationary process generating a system @xmath3 with the msj property , and let @xmath267 be an independent copy of @xmath268 .
we now define another stationary process @xmath38 by setting @xmath269 ( this means that @xmath270 tells us which are the two values taken by @xmath271 and @xmath272 , but neither which one is taken by @xmath271 nor which one is taken by @xmath272 . )
denote by @xmath7 the system generated by @xmath38 : @xmath7 is a factor of the cartesian product @xmath101 .
then @xmath3 and @xmath7 are certainly not disjoint , since the two processes @xmath13 and @xmath38 are clearly not independent of each other . however , suppose that @xmath7 and @xmath3 share a common factor which is not the system reduced to one point .
then , since @xmath3 has the msj property , this factor has to be @xmath3 itself .
therefore , we can find a third copy @xmath273 of @xmath13 , living in the cartesian square @xmath101 , and which is measurable with respect to the factor @xmath15-algebra generated by @xmath38 . by the 3-fold msj property of @xmath3 ,
either @xmath273 is equal to one of the processes @xmath13 or @xmath274 , possibly shifted by some constant integer @xmath83 , or @xmath273 is independent of @xmath275 .
the former case is impossible since @xmath38 , hence @xmath273 , is invariant by the `` flip '' @xmath276 , which is clearly not true for any shifted copy of @xmath13 or @xmath274 .
the latter case is also impossible since @xmath277 and @xmath278 generate the whole @xmath15-algebra in @xmath101 .
there exists however an important result which tells us how to derive some information on factors from the non - disjointness of two systems .
this theorem appeared for the first time in year 2000 , in two publications @xcite .
[ non - disjointness]if @xmath3 and @xmath7 are not disjoint , then @xmath7 has a non trivial common factor with some joining of a countable family of copies of @xmath3 .
start with a joining @xmath62 of @xmath0 and @xmath8 : then @xmath7 is a factor of @xmath68 . consider a countable family of copies of @xmath68 , and denote by @xmath279 their relatively independent joining over their common factor @xmath7 .
since this joining identifies all the projections on @xmath63 , it can be viewed as a probability measure on @xmath280 , and is characterized by @xmath281\cdots\ee_\lambda[\ind{a_k}|y]\,d\nu(y).\ ] ] observe that this probability measure is invariant under the shift on each @xmath159-fiber : @xmath282 moreover , @xmath279 conditioned on such a fiber is a product measure : the infinite direct product of @xmath62 conditioned on @xmath159 .
a relative version of kolmogorov 0 - 1 law ( precisely stated in @xcite , lemma 9 ) then tells us that the @xmath15-algebra of shift - invariant events coincides modulo @xmath279 with the @xmath15-algebra of @xmath159-measurable events .
now , let @xmath127 be a bounded measurable function defined on @xmath4 , which we also consider as a function on @xmath35 and on @xmath280 by setting @xmath283 , and @xmath284 .
applying birkhoff ergodic theorem to @xmath127 in the system defined by the shift and @xmath279 gives @xmath285\ = \
\ee_{\lambda}[f|y ] \quad \lambda_\infty\text{-a.e.}\ ] ] since @xmath3 and @xmath7 are not disjoint , we could have chosen @xmath286 , and then @xmath127 such that @xmath287 $ ] is not constant modulo @xmath62 .
then this conditional expectation generates in @xmath7 a non - trivial factor , which by can be identified to some factor of the joining of a countable family of copies of @xmath3 defined by the marginal of @xmath279 on @xmath288 .
we get from theorem [ non - disjointness ] some important disjointness results .
note that some properties of dynamical systems are stable under the operations of taking joinings and factors . we will call these properties _ stable properties_.
e.g. _ the case of the zero - entropy property : we know that a factor of a zero - entropy system still has zero entropy , and that any joining of zero - entropy systems also has zero entropy .
take now any @xmath289-system @xmath7 : we know that any non - trivial factor of @xmath7 has positive entropy , hence @xmath7 can not have a non - trivial common factor with a joining of copies of a zero - entropy @xmath3 .
the same argument also applies to the disjointness of discrete - spectrum systems with weakly mixing systems , since discrete spectrum is a stable property , and weakly mixing systems are characterized by the fact that they do not have any discrete - spectrum factor .
we thus have the following theorem : [ disjoint - classes - theorem ] any zero - entropy system is disjoint from any @xmath289-system .
any discrete - spectrum system is disjoint from any weakly mixing system .
the result of theorem [ non - disjointness ] leads to the introduction of a special class of factors when some dynamical system @xmath3 is given : for any other dynamical system @xmath7 , call _ @xmath3-factor _ of @xmath7 any common factor of @xmath7 with a joining of countably many copies of @xmath3 .
we also call _ @xmath3-factor @xmath15-algebra _ any factor @xmath15-algebra of @xmath7 associated to a @xmath3-factor of @xmath7 .
another way to state theorem [ non - disjointness ] is then the following : if @xmath7 and @xmath3 are not disjoint , then @xmath7 has a non - trivial @xmath3-factor .
in fact , the proof of this theorem gives a more precise result : for any joining @xmath62 of @xmath7 and @xmath3 , for any bounded measurable function @xmath127 on @xmath4 , the factor @xmath15-algebra of @xmath7 generated by the function @xmath290 $ ] is a @xmath3-factor @xmath15-algebra of @xmath7 . with the notion of @xmath3-factor , we can extend theorem [ non - disjointness ] in the following way , showing the existence of a special @xmath3-factor @xmath15-algebra of @xmath7 concentrating anything in @xmath7 which could lead to a non trivial joining between @xmath3 and @xmath7
. [ t - fact - max ] given two dynamical systems @xmath0 and @xmath8 , there always exists a maximum @xmath3-factor @xmath15-algebra of @xmath7 , denoted by @xmath291 . under any joining @xmath62 of @xmath3 and @xmath7 ,
the @xmath15-algebras @xmath292 and @xmath293 are independent conditionally to the @xmath15-algebra @xmath294 .
the proof of the theorem is based on the two following lemmas .
[ tfm1 ] let @xmath295 be a countable family of events in @xmath80 , such that for all @xmath296 , @xmath297 belongs to some @xmath3-factor @xmath15-algebra @xmath298 of @xmath7 . then there exists a @xmath3-factor @xmath15-algebra of @xmath7 containing all the @xmath297 s . for each @xmath299 , we have a joining @xmath300 of @xmath7 with a countable family @xmath301 of copies of @xmath3 , such that @xmath302 .
let us denote by @xmath303 the dynamical system defined by @xmath300 , and by @xmath62 the relatively independent joining of all the @xmath303 , @xmath299 , over their common factor @xmath7 .
we can see @xmath62 as a joining of @xmath7 with the countable family @xmath304 and , for each @xmath296 we have @xmath305 we conclude that the factor @xmath15-algebra of @xmath7 generated by all the @xmath298 s is a @xmath3-factor @xmath15-algebra , which certainly contains all the @xmath297 s .
[ tfm2 ] let @xmath49 be a factor @xmath15-algebra of @xmath7 .
if there exists a joining @xmath62 of @xmath3 and @xmath7 under which the @xmath15-algebras @xmath292 and @xmath293 are not independent conditionally to @xmath306 , then there exists a @xmath3-factor @xmath15-algebra @xmath307 of @xmath7 , not contained in @xmath49 .
the hypothesis of the lemma implies the existence of a bounded measurable function @xmath127 on @xmath4 such that , on a set of positive @xmath36-measure , @xmath308\ \not=\ \ee_{\lambda}[f(x ) | \f].\ ] ] the factor @xmath15-algebra @xmath307 of @xmath7 generated by the function @xmath309 $ ] is not contained in @xmath49 ; but we saw in the proof of theorem [ non - disjointness ] that @xmath307 is a @xmath3-factor @xmath15-algebra . in order to prove the existence of a maximum @xmath3-factor @xmath15-algebra , we define @xmath310 and we claim that it is a @xmath3-factor @xmath15-algebra . since @xmath156 is a lebesgue space , the @xmath15-algebra @xmath80 equipped with the metric @xmath311 is separable ( as usual , we identify subsets @xmath131 and @xmath150 of @xmath63 when @xmath312 )
. there exists a countable family @xmath295 dense in @xmath291 , and , thanks to lemma [ tfm1 ] , there exists a @xmath3-factor @xmath15-algebra @xmath49 containing all the @xmath297 s . by density
, we have @xmath313 but , since @xmath291 contains all the @xmath3-factor @xmath15-algebras , we have @xmath314 .
this proves the first assertion of theorem [ t - fact - max ] .
the second one is just the application of lemma [ tfm2 ] to @xmath291 .
the notion of @xmath3-factor was introduced in @xcite in order to study some aspects of the weak disjointness of dynamical systems .
it was used there to prove that if @xmath3 satisfies the property of being _ self - weakly disjoint to all order @xmath215 _ ( which is the natural generalization of self weak - disjointness to the case of @xmath83 copies of @xmath3 ) , then @xmath3 is weakly disjoint from _ any other dynamical system . _
the main open question in this area is whether the self weak disjointness of @xmath3 alone is sufficient to ensure this universal weak disjointness property . we know from theorem [ non - disjointness ] that if @xmath3 and @xmath7 are not disjoint , then the maximum @xmath3-factor of @xmath7 is not trivial .
one may ask whether the converse is true , but in fact it is not difficult to find a counterexample : indeed , take @xmath315 any stationary process taking its values in @xmath316 , and let @xmath317 be an i.i.d . process with @xmath318 taking each value 0 or 1 with probability @xmath319 , independent of @xmath320 . then , setting @xmath321 we get another i.i.d .
process @xmath322 with the same distribution as @xmath323 .
the @xmath320 process is clearly measurable with respect to the @xmath15-algebra generated by @xmath323 and @xmath322 , which means that the dynamical system @xmath7 generated by @xmath320 is a factor of a self - joining of the bernoulli shift generated by @xmath323 .
therefore , denoting by @xmath3 this bernoulli shift , we get that any dynamical system @xmath7 generated by a two - valued stationary process is a @xmath3-factor .
but we could have taken for @xmath320 any zero - entropy process , which is disjoint from any bernoulli shift ( theorem [ disjoint - classes - theorem ] ) .
however , we can notice that theorem [ non - disjointness ] gives a stronger conclusion when @xmath3 and @xmath7 are not disjoint .
indeed , this hypothesis being symmetric in @xmath7 and @xmath3 , we get in that case that both the maximum @xmath3-factor of @xmath7 and the maximum @xmath7-factor of @xmath3 are not trivial . in the counterexample given above , @xmath7 having zero entropy , the maximum @xmath7-factor of a bernoulli shift is trivial .
so we ask the following question : can we find two disjoint systems @xmath7 and @xmath3 such that both the maximum @xmath3-factor of @xmath7 and the maximum @xmath7-factor of @xmath3 are not trivial ? analyzing the proof of theorem [ t - fact - max ] , we can observe that the existence of the maximum @xmath3-factor @xmath15-algebra is in fact a corollary of the following : being a @xmath3 factor is a _
stable property _
( in the sense given in section [ stable - subsubsection ] ) . and indeed , instead of the notion of @xmath3-factor we could take any stable property and we would get by the same argument the existence of a maximum factor @xmath15-algebra satisfying this stable property .
some classical factor @xmath15-algebras are particular cases of this type : if we start from the stable property of having zero entropy , we get the _ pinsker @xmath15-algebra _ of the system , and if we start from the discrete - spectrum property , we get the _ kronecker factor @xmath15-algebra_.
we now turn to the relationships between the study of joinings and some ( weak and strong ) mixing properties .
we begin by a beautiful proof by ryzhikov of a well - known result concerning weak mixing .
there are many different ( but equivalent ! ) ways to define the property of weak mixing .
a very concise one is the following : we say that @xmath3 is weakly mixing if the cartesian square @xmath101 is ergodic .
it is a standard result in ergodic theory that this property implies the seemingly stronger one : [ weak - mixing - thm ] suppose that @xmath101 is ergodic .
then for any ergodic @xmath7 , @xmath37 is ergodic .
the classical proof of this implication makes use of the spectral theory of the action of the unitary operators @xmath324 and @xmath325 on the subspaces of @xmath137 and @xmath326 of functions with zero mean .
it can be shown that if @xmath3 and @xmath7 are ergodic , the cartesian square @xmath37 is ergodic if and only if @xmath327 and @xmath328 do not have any common eigenvalue .
we propose here an alternative approach by joinings to prove theorem [ weak - mixing - thm ] , due to ryzhikov @xcite .
let us start with a simple lemma on relatively independent joinings .
[ symmetrization - lemma ] let @xmath62 be a joining of @xmath3 and @xmath7 , and let @xmath329 be the relatively independent self - joining of @xmath68 above its factor @xmath7 .
if the marginal of @xmath329 on @xmath330 is the product measure @xmath207 , then @xmath62 is the product measure @xmath65 . for all @xmath103
, we have by definition of @xmath329 @xmath331\right)^2 \ , d\nu(y).\ ] ] thus , if the marginal on @xmath330 is the product measure , we get that for all @xmath103 , @xmath332\right)^2 \ , d\nu(y)\ = \
\left(\int_y\ee_\lambda[\ind{a}|y ] \ , d\nu(y)\right)^2,\ ] ] which is possible only if @xmath333 $ ] is a constant @xmath36-a.e .
, hence @xmath108 .
assuming that @xmath207 is an ergodic self - joining of @xmath3 , we are going to show that for all ergodic @xmath7 , @xmath334 , by checking that it is an extremal point in @xmath64 , and applying proposition [ characterization - of - ergodic - joinings ] .
suppose that for some @xmath335 and some joinings @xmath336 and @xmath337 of @xmath3 and @xmath7 , we have @xmath338 note that the relatively independent self - joining of @xmath65 above the factor @xmath7 is nothing but @xmath339 , which can be decomposed as follows : @xmath340 since @xmath207 is ergodic , each term of the rhs must have @xmath207 as marginal on @xmath330 .
but this implies that @xmath336 and @xmath337 satisfy the hypothesis of lemma [ symmetrization - lemma ] , hence both @xmath336 ad @xmath337 are the product measure @xmath65 .
strong mixing is also very easily characterized in term of joinings : by definition , the dynamical system @xmath3 is _ strongly mixing _ ( or simply _ mixing _ ) if @xmath341 , @xmath342 but this means precisely that the sequence @xmath343 converges in the topology of joinings to @xmath207 .
an elegant application of this characterization was found and used by ornstein in @xcite , where he constructed the first example of a transformation with no square root .
the dynamical system @xmath3 is mixing if and only if @xmath3 is weakly mixing and there exists some real number @xmath344 such that @xmath345 the fact that mixing implies weak mixing can be proved by the following argument using joinings : if @xmath3 is mixing , convergence of @xmath343 to @xmath346 also implies convergence of @xmath347 ( sequence of self - joinings of @xmath348 to @xmath349 . and
this gives that any @xmath101-invariant set in @xmath330 has measure zero or one , _
i.e. _ @xmath3 is weakly mixing .
condition is obviously necessary for @xmath3 to be mixing .
conversely , let us suppose that @xmath3 is weakly mixing and that holds .
then any cluster point @xmath62 of the sequence @xmath343 in @xmath85 satisfies @xmath350 , hence is absolutely continuous with respect to @xmath207 .
since @xmath3 is weakly mixing , @xmath207 is ergodic , therefore @xmath351 .
this proves that @xmath352 _
i.e. _ @xmath3 is mixing .
we finish this presentation by a long - standing open question in ergodic theory , and some of its relationships with joining theory .
recall that the property of ( strong ) mixing is also called _
2-fold mixing _ , because it involves only two subsets of the space .
a strengthening of this definition involving 3 ( or @xmath353 ) sets gives rise to the property of _ 3-fold mixing _
( respectively _ @xmath83-fold mixing _ ) : the dynamical system @xmath3 is said to be _ 3-fold mixing _
if @xmath354 , @xmath355 again , this definition can easily be translated into the language of joinings : @xmath3 is 3-fold mixing when the sequence @xmath356 converges in @xmath357 to @xmath358 as @xmath161 and @xmath359 go to infinity , where @xmath360 is the obvious generalization of @xmath206 to the case of self - joinings of order 3 .
whether 2-fold mixing implies 3-fold mixing was asked long ago by rohlin @xcite , and the question is still open today .
however , some important results have been established around this problem .
some of them could make us think that this implication is false , by exhibiting counterexamples to the same question but in a generalized context . in this category
, we must in particular cite ledrappier s example in @xcite of a @xmath361-action which is 2-fold but not 3-fold mixing .
more recently , it was noticed in @xcite that if we consider the property of being mixing relatively to a factor @xmath15-algebra ( roughly speaking , this means that we replace the measure by the conditional expectation with respect to this @xmath15-algebra ) , then we can find examples of 2-fold but not 3-fold mixing . on the other hand ,
there are also many results which go in the opposite direction . we know that 2-fold mixing implies 3-fold mixing in many cases : for example when @xmath3 has singular spectrum ( host , @xcite ) , when @xmath3 is rank one ( kalikow , @xcite ) or even for finite rank systems ( ryzhikov , @xcite ) , when @xmath3 is generated by a gaussian system ( leonov , @xcite ; see also totoki @xcite ) .
in fact , in many of these works it is shown that a stronger result holds for the respective class of systems under consideration , and this result concerns joining theory : for finite rank systems , as well as for singular spectrum systems , it is proved that any ergodic self - joining @xmath62 of order 3 which has pairwise independent marginals is necessarily the product measure . as far as gaussian systems are concerned
, this property has been established for an important subclass of the zero - entropy gaussian processes , called gag @xcite .
it is not known whether it can be extended to all zero - entropy gaussian processes .
more generally , we may ask the following important question : [ pairwise - independent - question ] does there exist a zero - entropy dynamical system @xmath3 and an ergodic joining @xmath362 for which the coordinates are pairwise independent but which is different from @xmath358 ? why a negative answer to this question would solve rohlin s multiple mixing problem is explained by the following remarks : first , if we had a 2-fold mixing but not 3-fold mixing dynamical system @xmath3 , the sequence @xmath356 would have a cluster point @xmath363 .
since @xmath3 is 2-fold mixing , the coordinates would be pairwise independent for @xmath62 .
and we could find an ergodic joining satisfying those two properties by taking an ergodic component of @xmath62 .
second , thouvenot proved that if we could find a counterexample to rohlin s question , then we could find one in the zero - entropy category .
this restriction to zero entropy is crucial since we can easily construct examples answering question [ pairwise - independent - question ] if we drop it : start from the @xmath364 bernoulli shift @xmath3 generated by the i.i.d .
process @xmath365 , taking its values in @xmath316 .
take an independent copy @xmath366 of this process and set @xmath367 then the three processes @xmath13 , @xmath274 and @xmath273 have the same distribution , are pairwise independent , but the self - joining of @xmath3 of order 3 that we get in this way is not the product measure .
observe also that a negative answer to question [ pairwise - independent - question ] would imply another important result in joining theory : it would show that 2-fold msj implies @xmath83-fold msj for all @xmath215 .
indeed , any 2-fold msj but not 3-fold msj would have an ergodic joining @xmath62 of order 3 for which the coordinates would be pairwise independent , but not independent .
the purpose of the following is to show that , if ever we could find some system satisfying the requirements of question [ pairwise - independent - question ] , it must be of a different nature than the construction of ledrappier for an action of @xmath361 , or than the relative example given in @xcite .
indeed , each of these two constructions leads to some `` joining '' @xmath62 , for which we get 3 pairwise independent coordinates @xmath368 , @xmath369 and @xmath370 , and for which the global independence is denied by the following property : we have some random variable @xmath13 defined on @xmath4 , taking its value in @xmath316 ( each value being taken with probability @xmath319 ) , and satisfying @xmath371 there are two ways of interpreting equation : first , the random variable @xmath13 takes its values in a finite alphabet @xmath5 , and there exists a function @xmath372 such that @xmath373 second , the random variable @xmath13 takes its values in a compact abelian group @xmath162 , and we have @xmath374 we want to show now that there can not exist an example answering question [ pairwise - independent - question ] of either of those types , by proving the following two propositions .
[ first - proposition]suppose that there exist some joining @xmath362 , with pairwise independent coordinates , and some random variable @xmath375 ( finite alphabet ) such that @xmath376 then * either the factor of @xmath3 generated by @xmath13 is periodic , * or this factor has an entropy at least @xmath377 .
[ second - proposition ] the dynamical system @xmath3 is assumed to be ergodic .
suppose that for some joining @xmath362 with pairwise independent marginals , there exists some measurable map @xmath378 ( compact abelian group ) which is not @xmath2-a.e .
constant , and such that @xmath379 then @xmath3 has positive entropy .
[ first - lemma ] let @xmath380 , @xmath381 and @xmath382 be 3 random variables taking their values in the finite alphabet @xmath5 , identically distributed , pairwise independent , and such that there exists a function @xmath372 satisfying @xmath383 then their common probability law is the uniform law on the set of values which are taken with positive probability .
taking a subset of @xmath5 instead of @xmath5 if necessary , we can assume that for all @xmath384 , @xmath385 for all @xmath296 and @xmath83 in @xmath5 , there exists a unique @xmath386 such that @xmath387 .
moreover , we have @xmath388 in other words , @xmath389 where @xmath390 and @xmath391 is the bistochastic matrix whose entries are @xmath392 . taking the limit as @xmath393 in the equality @xmath394
, we get that @xmath143 has to be the uniform law on @xmath5 . from the preceding lemma , we know that @xmath13 is uniformly distributed on the subset of values which are taken with positive probability .
but for any @xmath395 , lemma [ first - lemma ] also applies to @xmath396 let @xmath397 in @xmath5 be such that @xmath398 then , the conditional probability law of @xmath399 knowing the above event is uniform on the set of values to which it assigns a positive mass .
moreover , we easily check that the cardinal @xmath400 of the support of this conditional probability law does not depend on the choice of @xmath397 , and satisfies @xmath401 .
thus , * either the sequence @xmath402 reaches 1 , and in this case the factor generated by @xmath13 is periodic ; * or @xmath400 is always greater than or equal to 2 , and then the entropy of the factor generated by @xmath13 is at least @xmath377 .
[ second - lemma ] let @xmath380 , @xmath381 and @xmath382 be 3 random variables taking their values in the compact abelian group @xmath162 , identically distributed , pairwise independent , and satisfying @xmath403 then their common probability law is the haar measure on some compact subgroup of @xmath162 .
denote by @xmath36 the probability law of @xmath380 on @xmath162 . for @xmath36-almost
every @xmath404 , @xmath36 is invariant under the addition of @xmath129 , since knowing @xmath405 , both @xmath380 and @xmath406 are distributed according to @xmath36 .
let us define @xmath407 this is a closed subgroup of @xmath162 containing the support of @xmath36 .
thus @xmath36 is a probability law on @xmath408 , invariant under every translation on @xmath408 : @xmath36 is the haar measure on @xmath408 . let us consider @xmath409 defined by @xmath410 from lemma [ second - lemma ] , the probability law of @xmath411 is the haar measure on some compact subgroup @xmath408 of @xmath412 .
moreover , @xmath408 is invariant by the shift , denoted by @xmath15 , on @xmath412 .
the restriction of @xmath15 to @xmath408 is therefore an ergodic endomorphism of @xmath408 , which is a factor of @xmath3 ( this is the factor generated by @xmath13 ) .
a theorem by juzvinskii @xcite tells us that any such ergodic group endomorphism has to be a @xmath289-system ; hence @xmath3 has positive entropy .
observe that the two situations exposed in propositions [ first - proposition ] and [ second - proposition ] can be seen as particular cases of a third one , from which i do not know whether some similar conclusion can in general be derived .
what can be said about a dynamical system @xmath3 satisfying the following : there exists an ergodic self - joining @xmath362 , with pairwise independent marginals , and a map @xmath413 such that @xmath414
i want to express my special thanks to jean - paul thouvenot and valery ryzhikov , from whom i learnt most of what i know on joinings of dynamical systems ; to anthony quas for his help concerning propositions [ first - proposition ] and [ second - proposition ] ; to el houcein el abdalaoui , lise janvresse and yvan velenik for their support and advice in the preparation of this review .
thouvenot , _ some properties and applications of joinings in ergodic theory _ ,
ergodic theory and its connections with harmonic analysis ( alexandria , 1993 ) , london math .
lecture note ser . ,
205 , cambridge univ . press , cambridge , 1995 , pp . 207235 . | since their introduction by furstenberg @xcite , joinings have proved a very powerful tool in ergodic theory .
we present here some aspects of the use of joinings in the study of measurable dynamical systems , emphasizing on * the links between the existence of a non trivial common factor and the existence of a joining which is not the product measure , * how joinings can be employed to provide elegant proofs of classical results , * how joinings are involved in important questions of ergodic theory , such as pointwise convergence or rohlin s multiple mixing problem . |
2 the two - dimensional classical @xmath0 model exhibits a non standard behaviour . long - range order of spins with continuous symmetries is indeed forbidden according to the mermin - wagner theorem @xcite , and thus there is no spontaneous magnetisation at finite temperature .
the transition is governed by unbinding of topological defects .
the low - temperature regime was investigated by berezinskii @xcite and the mechanism of unbinding of vortices was studied by kosterlitz and thouless @xcite using renormalization group methods . for reviews , see e.g. refs .
@xcite .
the notations are specified below : we consider a @xmath5 @xmath0-model on a square lattice with two - components spin variables @xmath6 located at the sites @xmath7 of a lattice @xmath8 of linear extent @xmath9 .
the spins interact through the usual nearest - neighbour ferromagnetic interaction -=k_w__w_w+ , [ ham ] where @xmath10 and @xmath11 is a unit vector in the @xmath12direction . at low temperature ,
the system is partially ordered , apart from the existence of vortices , which appear in pairs in increasing number with increasing temperature . in the low - temperature limit ,
the effect of vortices can be neglected and the behaviour is governed by spin - wave excitations , obtained after expanding the cosine .
this harmonic approximation is justified , provided that the spin disorientation remains small , i.e. at sufficiently low temperature . within this approximation
, the quadratic energy in the boltzmann factor leads to a gaussian equilibrium distribution and the two - point correlation function becomes @xcite _ w_1_w_2|w_1-w_2|^-1/2k , hence ( t)=k_bt/2j.[eta - sw ] when the temperature increases , the influence of vortices becomes more prominent , producing a deviation from the linear spin - wave contribution in equation ( [ eta - sw ] ) , but the order parameter correlation function still decays algebraically with an exponent @xmath13 which depends on the temperature .
the existence of such a scale - invariant power - law decay of the correlation function is the signature of a continuous line of fixed points at low temperatures . in this paper
we are interested in the end point of this critical line where the topological transition takes place : at the transition temperature @xmath14 ( usually called berezinskii - kosterlitz - thouless critical temperature ) , the pairs are broken and the system becomes completely disordered .
this very peculiar topological transition is characterised by essential singularities when approaching the critical point from the high temperature phase , and at @xmath14 the correlation function exponent takes the value @xmath3 @xcite . in order to recover this result
, many numerical studies were performed in the last decade using monte carlo simulations ( see e.g. refs .
@xcite ) , but the analysis was made difficult by the existence of logarithmic corrections , e.g. _
l|w_12|^-(|w_12|)^/2 , [ eq9 ] at @xmath14 @xcite . due to these logarithmic corrections which make the fits quite difficult ,
the value of @xmath1 at @xmath14 was a bit controversial as shown in table 1 of reference @xcite .
the resort to large - scale simulations , up to systems as large as @xmath15 , was then needed in order to confirm this picture @xcite .
recently , we proposed a rather different approach @xcite which is easily implemented .
we use the covariance law of correlation functions under the mapping of a two - dimensional system confined inside a square onto the half - infinite plane .
the scaling dimensions are then obtained through a simple power - law fit when the correlators are expressed as functions of conveniently rescaled variables .
the shape effects are explicitly encoded in the conformal mapping .
this is the crucial point , since then even very small systems lead to promising results .
we may then determine accurately the bulk correlation function exponent @xmath1 at @xmath14 but also , by choosing convenient boundary conditions , the surface critical exponent which describes the decay of the correlation function parallel to a free surface , @xmath2 .
2 for a scale - invariant system at its critical point ( which also exhibits the properties of isotropic scaling and short - range interactions ) , conformal invariance enable to include explicitly the shape dependence in the functional expression of the correlators through the conformal covariance transformation under a mapping @xmath16 : _
w_1_w_2=|w(z_1)|^-x_|w(z_2)|^-x _ _ z_1_z_2 . [ covconf ] the functional expression of the correlation function inside the square geometry @xmath7 simply follows the mapping @xmath16 which realizes the conformal transformation of the half - plane @xmath17 ( @xmath18 ) inside a square @xmath19 of size @xmath20 ( @xmath21 , @xmath22 ) with free boundary conditions along the four edges .
this is realized by a schwarz - christoffel transformation @xcite w(z)=l2kf(z , k ) , z = sn(2kwl ) .
[ eq - schchr ] here , @xmath23 is the elliptic integral of the first kind , @xmath24 the jacobian elliptic sine ,
@xmath25 the complete elliptic integral of the first kind , and the modulus @xmath26 depends on the aspect ratio of @xmath8 and is here solution of @xmath27 . using the mapping ( [ eq - schchr ] ) ,
one obtains the local rescaling factor in equation ( [ covconf ] ) , @xmath28^{-1/2}$ ] .
one simply has now to include in equation ( [ covconf ] ) the expected behaviour of @xmath29 in the half - infinite geometry in order to get the functional expression of the corresponding quantity in the confined system .
for example in the semi - infinite geometry @xmath17 ( the free surface being defined by the @xmath30 axis ) , the two - point correlation function is fixed up to an unknown scaling function : fixing one point @xmath31 close to the free surface ( @xmath32 ) of the half - infinite plane , and leaving the second point @xmath33 explore the rest of the lattice , the following behaviour is expected : _
z_1_z_2 _ ~(y_1-y_2)^-x_(),[eq : g ] with @xmath34 .
the dependence on @xmath35 of the universal scaling function @xmath36 is constrained by the special conformal transformation @xcite , and its asymptotic behaviour is implied by scaling e.g. @xmath37 when @xmath38 , with @xmath39 the surface scaling dimension . keeping @xmath40 fixed inside the square , the two - point correlation function becomes @xmath41^{-x_\sigma } \psi(\omega)\nonumber\\ & & \kappa(w)= { \rm im}\left[{\rm sn}\frac{2kw}{l } \right]\times \left| \left(1-{\rm sn}^2\frac{2kw}{l } \right ) \left ( 1-k^2{\rm sn}^2\frac{2kw}{l } \right)\right|^{-1/2 } , \label{eqkappa}\end{aligned}\ ] ] 2 where @xmath42 specifies that the boundary conditions ( bcs ) are chosen free along the edges of the square .
this expression is correct up to a constant amplitude determined by @xmath43 which is kept fixed , but the function @xmath44 is still varying with the location of the second point , @xmath7 .
this makes this expression not so useful in practice ( see e.g. reference @xcite for an application ) . in order to cancel the role of the unknown scaling function , it is more convenient to work with a density profile @xmath45 in the presence of symmetry breaking surface fields @xmath46 on some parts of the boundary @xmath47 of the lattice @xmath8 . in the case of fixed - free bcs denoted by @xmath48 ( the spins located on the @xmath49 half axis
are kept fixed while those of the @xmath50 are free ) , burkhardt and xue have shown in the case of ising and potts models that in the half - plane the order parameter profile obeys the following expression @xcite : m_+f(z)=y^-/2()^_/2 . in the square geometry , the fixed - free bcs correspond to keeping the spins fixed for @xmath51 , and the corresponding profile is expected to obey the following ansatz : @xmath52^{-x_\sigma}[\mu(w)]^{x_\sigma^1 } , \nonumber\\ % & & \kappa(w)= { \rm im}\left[{\rm sn}\frac{2kw}{l } \right]\times % \left| \left(1-{\rm sn}^2\frac{2kw}{l } % \right ) \left ( 1-k^2{\rm sn}^2\frac{2kw}{l } \right)\right|^{-1/2},\nonumber\\ & & \mu(w)= \frac { 1}{\sqrt 2}\left(1+{\rm re}\left[{\rm sn}\frac{2kw}{l } \right]\times \left|\ \!{\rm sn}\frac{2kw}{l}\right|^{-1}\right)^{1/2 } \label{eqkappamu}\end{aligned}\ ] ] 2 where @xmath53 was defined in equation ( [ eqkappa ] ) .
in this paper we will essentially apply equation ( [ eqkappamu ] ) in order to determine the critical exponents @xmath1 and @xmath2 at the kt point . in ref .
@xcite , the bulk exponent was deduced from a similar analysis of the order parameter profiles in a square - shaped system with fully fixed bcs .
2 the application of the simple power - law in equation ( [ eqkappamu ] ) requires a relatively precise numerical determination of the order parameter profile of the @xmath5 @xmath0 model confined inside a square with mixed fixed - free boundary conditions . fixing the spins along some part @xmath47 of the boundary , e.g. @xmath54 , plays the role of ordering surface fields @xmath55 .
the resort to cluster update algorithms ( wolff algorithm here ) is necessary in order to prevent the critical slowing down : the central idea is the identification of clusters of sites using bond variables connected to the spin configuration .
the spins of the clusters are then updated and the algorithm is particularly efficient if the percolation threshold of the bond process occurs at the transition temperature of the spin model .
this ensures the updating of clusters of all sizes in a single mc sweep . for the @xmath0 model , ising variables defined in the wolff algorithm by the sign of the projection of the spin variables along some random direction .
the bonds are then introduced through the kasteleyn - fortuin random graph representation .
the percolation threshold for these bonds coincides with the kosterlitz - thouless point @xcite , which guarantees the efficiency of the wolff cluster updating scheme @xcite at @xmath14 .
when one uses particular boundary conditions , with fixed spins along some surface as required here , the wolff algorithm should become less efficient , since close to criticality the unique cluster will often reach the boundary and no update is made in this case . to prevent this
, we use the symmetry properties of the hamiltonian ( [ ham ] ) : even when the cluster reaches the fixed boundaries , it is updated , and the order parameter profile is then measured with respect to the new direction of the boundary spins , @xmath56 .
the new configuration reached has the right statistical weight . using this procedure ,
we discard @xmath57 sweeps for thermalization , and the measurements are performed on @xmath57 production sweeps . the simulations at @xmath14 on a typical system of size @xmath58 takes a few hours on a standard pc with 733 mhz processor , to get a smooth profile as shown in figure 1 .
= 4.5 cm * figure 1 : * mc simulations of the @xmath5 @xmath0 model inside a square @xmath59 spins with mixed fixed - free boundary conditions at the kt transition temperature @xmath60 ( average over @xmath57 mcs / spin after cancellation of @xmath57 for thermalization ) .
2 in ref .
@xcite , we have shown that the logarithmic corrections at the kt point are negligible for the order parameter profile with fixed bcs . assuming that there will be no substantial difference for fixed - free bcs
, we perform a fit of @xmath62 against the two - dimensional linear expression , @xmath63 , as expected from equation ( [ eqkappamu ] ) . the value obtained for the bulk correlation function exponent
@xmath1 will be a test for the absence of logarithmic corrections .
the simulations are performed at the value @xmath64 found in the literature , e.g. in ref .
the fit leads to accurate determinations of both exponents for different system sizes ranging between @xmath65 and @xmath66 . although it would easily be possible to produce simulations for larger systems
, there is no need here , since the results are already quite stable as shown in table 1 and in figure 2 .
this is fully coherent with the result of kosterlitz and thouless rg analysis @xmath67 .
we can thus also consider as reliable the value of the surface exponent @xmath68 .
= 7.2 cm * figure 2 : * size dependence of the bulk and surface scaling dimensions at the kt transition plotted as a function of the inverse linear size of the lattice .
2 equation ( [ eqkappa ] ) provides a test in order to control the quality of the surface exponent obtained above . indeed , if one plots @xmath70^{x_\sigma}$ ] _ vs _ @xmath71
, one should obtain the universal scaling function @xmath44 whose asymptotic large distance behaviour should lead to a consistent surface exponent through the expected power - law behaviour @xmath72 .
= 7.2 cm * figure 3 : * universal scaling function of the surface - volume correlation function for @xmath73 .
the power - law behaviour @xmath37 shown in dotted line agrees with a value @xmath74 .
the scaling function @xmath44 is shown in figure 3 .
it exhibits a clear power - law behaviour consistent with the previous value of the surface scaling dimension .
we mention here that the data were obtained for both quantities , @xmath69 and @xmath61 , with completely independent runs , since the boundary conditions are different in equations ( [ eqkappa ] ) and ( [ eqkappamu ] ) and the quantities themselves are different ( 2-point or 1-point correlators ) .
the data collapse is quite good and supports the validity of equation ( [ eqkappa ] ) .
2 in this paper we have performed standard mc simulations of the @xmath5 @xmath0 model at the kosterlitz - thouless topological transition .
combined to the use of a conformal mapping , they provide a simple numerical determination of bulk and surface critical exponents associated to the algebraic decay of the correlation function .
the well known result of kosterlitz and thouless for the bulk exponent is recovered and as far as the author knows this is the first determination of the surface exponent @xmath75 . on the contrary to the bulk case , @xmath2 is different from the ising value ( which is equal to @xmath76 ) @xcite .
99 ; . .
; . ; d.r .
nelson , in phase transitions and critical phenomena , ed . by c. domb j.l .
lebowitz , academic press , london 1983 , p. 1 .
, statistical field theory , cambridge university press , cambridge 1989 , vol .
1 . ; . ; ; . .
, e - print cond - mat/0208521 . ,
conformal invariance and critical phenomena , springer , heidelberg 1999 .
, mthodes de la thorie des fonctions dune variable complexe , mir , moscou 1972 , chap .
burkhard and xue gave several predictions for conformally invariant profiles in strip geometries @xmath77 or in their half - infinite counterpart @xmath17 : considering a semi - infinite system described by @xmath78 , @xmath79 , ordinary scaling implies a functional form in the half - plane , @xmath80 , where @xmath81 specifies the boundary conditions . under the logarithmic transformation into a strip of width @xmath9 , @xmath82 , one obtains the order parameter profile as @xmath83^{-x_\sigma } f_{ab}\left ( \frac{\pi l}{l}\right ) $ ] . .
k. binder , in phase transitions and critical phenomena , edited by c. domb and j.l .
lebowitz ( academic press , london , 1983),vol . | monte carlo simulations of the two - dimensional @xmath0 model are performed in a square geometry with free and mixed fixed - free boundary conditions . using a schwarz - christoffel conformal mapping
, we deduce the exponent @xmath1 of the order parameter correlation function and its surface equivalent @xmath2 at the kosterlitz - thouless transition temperature .
the well known value @xmath3 is easily recovered even with systems of relatively small sizes , since the shape effects are encoded in the conformal mapping .
the exponent associated to the surface correlations is similarly obtained @xmath4 .
epsf # 1#2#3#4#5#1 , # 2 * # 3 * ( # 5 ) # 4 |
this work concerns the inverse scattering problem of recovering an impenetrable obstacle by the corresponding acoustic wave detection .
the problem has its physical origin in radar / sonar , geophical exploration , non - destructive testing and medical imaging ( cf .
@xcite ) . let @xmath2 be a bounded lipschitz domain in @xmath0 , @xmath1 , such that @xmath3 is connected .
@xmath2 represents an impenetrable obstacle located in the space and it is assumed to be unknown / inaccessible . in order to identify @xmath2 , one sends a time - harmonic detecting plane wave of the form @xmath4 which is an entire solution to the helmholtz equation @xmath5 in @xmath0 .
the presence of the obstacle @xmath2 interrupts the propagation of the plane wave , leading to the so - called scattered wave field @xmath6 , which exists only in the exterior of the obstacle .
the total wave field @xmath7 satisfies the following helmholtz system @xmath8 in , @xmath9 or @xmath10 , with @xmath11 denoting the exterior unit normal vector to @xmath12 , corresponding to that @xmath2 is sound - soft or sound - hard , respectively . in the following , we set @xmath13 and @xmath14 , denoting the wavenumber and incident direction of the plane wave , respectively .
the pde system is well - understood with @xmath15 possessing the following asymptotic expansion ( cf .
@xcite ) @xmath16 which holds uniformly in @xmath17 , where @xmath18 and @xmath19 .
@xmath20 is known as the far - field pattern and we shall write @xmath21 to specify its dependence on the observation direction @xmath22 , wavenumber @xmath23 and incident direction @xmath24 , as well as the obstacle @xmath2 .
@xmath20 is real - analytic in @xmath22 , and hence if it is known on any open patch of @xmath25 , then it is known on the whole sphere by the analytic continuation ; see @xcite .
the inverse problem that we are concerned with in the present paper is to recover @xmath2 by knowledge of @xmath26 , which is known to be nonlinear and ill - posed ( cf .
it is noted that the inverse problem is formally posed with a fixed @xmath27 and all @xmath28 .
hence , there is a widespread belief that one can recover @xmath2 by using the far - field pattern corresponding to a single incident plane wave @xmath29 , which is referred to as a single far - field measurement .
however , this still remains to be a longstanding problem with very limited progresses in the literature .
if the obstacle is of small size ; roughly speaking , smaller than half of the detecting wavelength , the unique recovery result was established in @xcite .
if the obstacle is extremely rough " in the sense that its boundary is nowhere analytic , the unique recovery result was established in @xcite .
if the obstacle is of general polyhedral type , the corresponding study is almost exclusive @xcite .
recently , some qualitative numerical schemes of recovering the obstacles by a single far - field measurement were proposed in @xcite , where certain restrictive a priori assumptions have to be imposed on the obstacles .
another challenging issue in the study of inverse scattering problems is the recovery by phaseless data , say the modulus of the far - field pattern , @xmath30 .
the only recent result we are aware of is @xcite , where the unique determination of a scattering potential by the phaseless far - field measurements was established . in this paper
, we develop a novel scheme of recovering a polyhedral obstacle by using only a few high - frequency far - field measurements .
the obstacle could be sound - soft or sound - hard .
the crux is the observation that the modulus of the far - field pattern in the backscattering aperture possesses a certain local maximum behavior , from which one can determine the exterior normal directions of the front sides / faces . then by using the modulus of the backscattering data corresponding a few incident plane waves with suitably chosen incident directions , one can determine the exterior unit normal vector of each side / face of the obstacle .
after the determination of the exterior unit normals , the recovery is reduced to a finite - dimensional problem of determining a location point of the obstacle and the distance of each side / face away from the location point . for the latter reconstruction , the far - field data with phases would be used .
our study is based on the high - frequency asymptotics , namely the kirchhoff or the physical optics approximation .
however , our numerical experiments show that the high - frequency requirement could be relaxed to a certain extent .
moreover , in order to simplify the discussion , we focus on convex polyhedral obstacles in the present study .
however , through our theoretical arguments , it can be expected that the method developed would work for non - convex obstacles , but under certain geometrical constraints .
we focus on developing the novel inverse scattering scheme for convex obstacles in the present study and leave the technical and tedious derivation of the geometrical conditions for non - convex obstacles for a forthcoming work .
the rest of the paper is organized as follows . in section 2 , we consider the physical optics approximation on the high - frequency scattering from an admissible polyhedral obstacle , and derive the local maximum behavior of the modulus of the corresponding far - field pattern . in section 3
, we present the recovery scheme in detail .
section 4 is devoted to numerical experiments to validate the applicability and effectiveness of the proposed method and the paper is concluded in section 5 with some discussion .
throughout the present section , we let @xmath31 and @xmath32 be fixed . let @xmath2 be a convex polygon in @xmath33 or a convex polyhedron in @xmath34 , such that @xmath35 where each @xmath36 represents an open side / face of @xmath12 , and shall be referred to as a _ cell _ in what follows . in the sequel , @xmath2
is referred to as a polyhedral obstacle .
let @xmath37 , @xmath38 denote the exterior unit normal vector to @xmath12 , and we set @xmath39 clearly , @xmath40 is a constant unit vector .
define @xmath41 to be , respectively , the back - face and front - face of @xmath12 with respect to the incident direction @xmath24 .
let @xmath42 , @xmath43 and @xmath44 be fixed a priori positive constants .
it is further assumed that @xmath45 roughly speaking , implies that the obstacle should not be very round " or
sharp " , and a generic condition which can guarantee this assumption is that the angle between any two adjacent cells is bounded below and above by certain constants ( depending on the obstacle ) .
assumption ( i ) means that we are considering the scattering in the high - frequency regime .
if a polyhedral obstacle @xmath2 satisfies the above three assumptions , then it is called an _ admissible _ obstacle with respect to the incident plane wave @xmath46 .
denote @xmath47 the forward - scattering and backscattering apertures , respectively .
let @xmath48 be a front - cell of @xmath12 , and @xmath49 denote its unit normal vector pointing to the exterior of @xmath2 .
define @xmath50 satisfying @xmath51 to be the critical observation direction with respect to @xmath24 and @xmath40 ; see fig . 1 for a 2d illustration .
we note that one clearly has @xmath52 .
it is directly calculated that the critical direction is given by @xmath53 on the other hand , for the subsequent use , we note that by using and the fact that @xmath54 , one has by innerly producting both sides of with @xmath24 @xmath55 hence , by combining and , one further has
@xmath56 2d illustration of the relation between the incident direction @xmath24 , the exterior unit normal vector @xmath40 and the critical observation direction @xmath57 . , scaledwidth=50.0% ] ( a ) plot of the real part of the scattered wave @xmath58 in the vicinity of a sound - soft triangular obstacle , and ( b ) polar graph of the square power of the phaseless far - field data @xmath59 in polar coordinates corresponding to the triangle due to an incident plane wave @xmath46 with @xmath60 .
the black arrow indicates both the incident and the forward - scattering directions , while the two red arrows represents the two critical observation directions .
the green arrows indicates the exterior directions normal to the sides of the triangular obstacle . , title="fig:",scaledwidth=48.0% ] ( a ) plot of the real part of the scattered wave @xmath58 in the vicinity of a sound - soft triangular obstacle , and ( b ) polar graph of the square power of the phaseless far - field data @xmath59 in polar coordinates corresponding to the triangle due to an incident plane wave @xmath46 with @xmath60 .
the black arrow indicates both the incident and the forward - scattering directions , while the two red arrows represents the two critical observation directions .
the green arrows indicates the exterior directions normal to the sides of the triangular obstacle .
, title="fig:",scaledwidth=48.0% ] ( a)(b ) next , we present the major result motivating the recovery scheme that we are going to develop in the next section . before that , we give a numerical example by plotting the real part of the scattered wave field and the square power of the modulus of the associated far - field pattern corresponding to a sound - soft triangle due to an incident plane wave impinging from southeast to northwest ; see fig .
[ fig:2 ] for the illustration .
clearly , one can see a certain local maximum behavior from fig .
[ fig:2](b ) . the observation direction ( black arrow ) associated with the largest magnitude of phaseless data points to the incident / forward - scattering direction , which gives no information of the obstacle . while the other two critical observation directions ( red arrows ) associated with the second and third largest magnitude of phaseless data provide profound information of unknown obstacle , which signify the major reflection angles due to the physical optics approximation in the high frequency scattering .
the exterior unit normal direction can thus be determined by connecting the arrow heads of incident and critical observation directions by yellow lines . those unit vector ( green arrows ) parallel to the yellow lines are the desired normal directions of those sides of the triangular obstacle in the backscattering aperture as shown in fig .
[ fig:2](b ) .
we shall present a rigorous mathematical justification for this local maximum behavior , which also forms the basis of our subsequent recovery scheme .
we first give some discussion on the kirchhoff or the physical optics approximation ( cf .
@xcite ) , which plays a key role in our recovery scheme . considering the present study
, it says that for the scattering from a convex polyhedral obstacle due to a high - frequency plane wave , the wave field near the boundary of the obstacle is composed of two parts : the contribution of the incident and reflected waves where they are present , and the contribution of the diffraction due to the corners and/or the edges of the obstacle .
the kirchhoff or the physical optics approximation takes the first contribution as the total wave field near the boundary of the obstacle .
however , we would like to remark the study on rigorously justifying such approximation is still not fully understood and we refer to @xcite for an excellent account on the existing progress in the literature . for our study , we assume that the physical optics approximation holds true . throughout the rest of the paper , we let a cell @xmath61 be parameterized as @xmath62 where @xmath63 is the unit normal vector to @xmath61 pointing to the exterior of the obstacle , and @xmath64 denotes the distance between the origin and the line / plane containing @xmath61 .
let @xmath65 denote the affine cell of @xmath61 defined by @xmath66 in the sequel , we let @xmath67 denote the euclidean reflection with respect to @xmath65 . now
, we consider the scattering locally near a boundary cell , say @xmath36 due to a plane wave @xmath46 .
let @xmath68 where @xmath69 is a fixed point , and @xmath70 .
it is easily verified that @xmath71 satisfies @xmath72 .
moreover , one can further verify that @xmath73 @xmath71 is the reflected wave field of @xmath74 with respect to the cell @xmath36 . using the physical optics approximation , we have [ thm : ka ] let @xmath2 be an admissible polyhedral obstacle with respect to the plane wave field @xmath75 , as described in . using the physical optics approximation , one has @xmath76 if @xmath2 is sound - soft ; and @xmath77 if @xmath2 is sound - hard . according to our earlier discussion on the physical optics approximation
, one takes @xmath78 if @xmath2 is sound - soft .
then by using , one directly verifies for @xmath79 , whereas if @xmath80 , it is not illuminated , where one then takes @xmath81 . if @xmath2 is sound - hard , then one takes @xmath82 which readily verifies .
it is emphasized again that theorem [ thm : ka ] is mainly based on physical observation , and its rigorous justification still largely remains open in the literature .
nevertheless , our subsequent study on the inverse scattering problem , along with the corresponding numerical experiments , validates such approximation as well . in the sequel
, we let @xmath83 where @xmath84 denotes the zeroth - order hankel function of the first kind .
@xmath85 is the fundamental solution to @xmath86 .
the following lemma shall be needed and its proof can be found in @xcite .
[ lem:1 ] for the scattering of a plane wave field @xmath87 in from an obstacle @xmath2 , we have @xmath88 and @xmath89,\ ] ] where the dimensional parameter @xmath90 is given by @xmath91 [ lem:2 ] using the physical optics approximation in lemma [ thm : ka ] , for the scattering of a plane wave @xmath87 in from an admissible polyhedral obstacle @xmath2 , one has that if @xmath2 is sound - soft @xmath92 whereas if @xmath2 is sound - hard @xmath93 this is a straightforward consequence of in lemma [ lem:1 ] and , and in lemma [ thm : ka ] . by using lemma [ lem:2 ] , we are now in a position to present the local maximum behavior of @xmath94 .
[ thm:1 ] let @xmath2 be an admissible sound - soft or sound - hard polyhedral obstacle with respect to the incident plane wave @xmath46 as described earlier .
suppose that @xmath48 is a front cell of the obstacle , and @xmath40 is the unit normal vector to @xmath36 pointing to the exterior of @xmath2 , @xmath95 .
let @xmath50 be the critical observation direction with respect to @xmath24 and @xmath40 .
under the physical optics approximation , @xmath57 is a local maximum point of @xmath96 .
we first consider the case that @xmath2 is an admissible sound - soft polyhedral obstacle . by in lemma [ lem:2 ] , we have @xmath97\\ = & \widetilde{\gamma}(n , k)\cdot [ i_1(\hat x)+i_2(\hat x ) ] , \end{split}\ ] ] where @xmath98 and @xmath99 we next analyze the behavior of @xmath100 , @xmath101 , in a small neighborhood of @xmath102 on @xmath25 , say @xmath103 , @xmath95 .
since @xmath104 by assumption , we see that there exists a non - asymptotic constant @xmath105 such that @xmath106 represents a unit tangent vector on @xmath12 .
hence , by and , one has by direct calculations that @xmath107 , we have @xmath108 where @xmath109 is any fixed point on @xmath36 . since @xmath110 one clearly sees that @xmath111 achieves its local maximum value at @xmath102 , which in combination with completes the proof of the theorem for the sound - soft case .
the sound - hard case can be shown by following a similar argument . by using
, one has @xmath112 , \end{split}\ ] ] where @xmath113 we consider the behavior of @xmath114 , @xmath101 , in a small neighborhood of @xmath102 on @xmath25 , say @xmath115 , @xmath95 .
following a similar argument to , together with the fact , one can see that @xmath116 achieves its local maximum value at @xmath102 . on the other hand , by a similar argument in deriving
, one has that latexmath:[\[\label{eq : kkk0 } hence , @xmath102 is local maximum point of @xmath96 .
the proof is complete . in theorem [ thm:1 ] , we only consider the local maximum behavior of @xmath30 in the backscattering aperture .
in fact , by , one clearly sees that @xmath30 achieves its ( global ) maximum value at @xmath118 .
however , this maximum behavior in the forward - scattering aperture gives us no information about the obstacle , which is demonstrated and verified in fig .
[ fig:2 ] .
it is noted from that for a sound - hard obstacle if @xmath119 , then both @xmath120 and @xmath121 vanish , and hence such an @xmath22 should be a local minimum point of @xmath122 .
however , it can not be guaranteed that such an @xmath22 belongs to the backscattering aperture .
nevertheless , it could be used in combination with the critical observation direction as a double indicator .
based on our study in the previous section , we shall present a recovery scheme of identifying an admissible polyhedral obstacle @xmath2 . to that end , we let @xmath23 be a sufficiently large wavenumber such that assumption [ eq : cond1 ] is fulfilled . then , let @xmath123 , @xmath124 be a few properly chosen incident directions .
the basic requirement is that the union of the front - faces with respect to @xmath123 , @xmath124 should cover the whole boundary @xmath12 .
furthermore , we shall assume that @xmath2 is an admissible obstacle with respect to each @xmath123 .
as remarked earlier , this assumption is generically satisfied by an overall not very round " obstacle .
first of all , clearly , by theorem [ thm:1 ] and the discussion following its proof , one can recover a family of exterior unit normal vectors .
however , it must be emphasized that multiple groups of normal directions are actually determined corresponding to detecting waves with different incident directions . within each group ,
those normal directions point roughly to the same direction .
up to this stage , we choose the normal direction , which is associated with the critical observation angle with the largest magnitude of phaseless data within the group , to be an _ effective _ exterior unit normal direction . following this principle
, we can recover all the exterior unit normal vector @xmath40 to each of the cell @xmath36 to @xmath12 , @xmath125 . in turn
, we also know the number of cells of the obstacle .
next , we proceed to identify each cell , and we let @xmath126 be a fixed location point .
let each @xmath36 be parameterized as follows @xmath127 where @xmath128 denotes the distance between the origin and the line / plane containing the cell @xmath129 .
we note that @xmath40 and @xmath128 , @xmath125 , and @xmath130 uniquely determine the convex obstacle @xmath2 .
it is highlighted that the location point @xmath130 can be initially guessed by scheme i in @xcite using single - shot measurement data at low frequency .
hence , after the determination of @xmath130 and @xmath40 , @xmath125 , the corresponding inverse problem reduces to a finite dimensional problem of finding @xmath128 , @xmath125 .
but we have richer data set @xmath131 to that purpose .
now , let us take the sound - soft obstacle as an illustration to derive the rest of the recovery scheme .
suppose @xmath132 is a critical observation direction corresponding to @xmath36 with respect to a certain @xmath133 , @xmath134 .
then , by , we have @xmath135 eq .
is clearly a finite dimensional nonlinear problem with respect to @xmath128 s , which are hidden in the implicit functions defining @xmath36 s . by supplementing @xmath136
particularly chosen normal directions and their respective critical observation angles , one can arrive at a nonlinear system of @xmath136 equations with @xmath136 unknown @xmath128 , @xmath125 .
the cells can be determined all at once by solving the finite dimensional problem .
it is emphasized that the chosen @xmath136 nonlinear equations hold only approximately , thus we convert it into a nonlinear least - squares minimization problem to yield a more stable and accurate identification .
now we are in a position to present our main reconstruction algorithm of polygonal obstacles which is sketched in fig .
[ fig:11 ] .
employ scheme i proposed in @xcite to locate the position @xmath130 of the polygonal scatterer using an incident detecting wave of low frequency .
determine the @xmath136 effective exterior normal vectors @xmath40 , @xmath137 and their associated critical observation angles from multiple groups of candidate normal vectors associated with all the critical observation angles . step 3 .
given the distances @xmath128 s from @xmath130 along the direction @xmath40 , @xmath137 , we can locate all the perpendicular points @xmath138 s .
extend the line / plane at @xmath138 s perpendicular to the respective @xmath40 and denote those crossing points to be vertices @xmath139 s . the line / plane of the polygonal scatterer can be determined by @xmath140 where @xmath141 if @xmath142 and @xmath143 if @xmath144 .
select the first @xmath136 critical observation angles @xmath145 , @xmath146 from all the @xmath132 , @xmath147 , @xmath137 according to the decreasing order of magnitude of local backscattering maxima and convert the corresponding @xmath136 nonlinear equations associated with with @xmath136 unknowns @xmath128 s into a least - squares minimization problem .
it is noted that the integral to the right hand side of is approximated by the trapezoidal quadrature rule with sufficiently fine step size .
@xmath148 step 6 .
the minimum of the multivariable nonlinear least - squares minimization problem is obtained by employing a derivative - free trust region method via a local quadratic surrogate model - based search algorithm .
interested readers may refer to @xcite and the references therein .
sketch of the recovery scheme .
, scaledwidth=60.0% ]
in this section , we present some numerical tests to verify the applicability and effectiveness of the proposed recovery scheme in two dimensions . in the sequel , the forward equation ( [ eq : helm ] ) is first solved by using the quadratic finite element discretization on a truncated circular domain enclosed by a pml layer .
the forward solver is iterated on a sequence of successively refined meshes till the relative error of two successive finite element solutions between the two adjacent meshes is below @xmath149 .
then the scattered data are transformed into the far - field data by employing the kirchhoff integral formula on a closed circle ( 2d ) enclosing the scatterer .
first of all , let s fix the parameter settings .
for the positioning purpose , we apply scheme i ( see @xcite ) with @xmath150 to detect the initial displacement guess @xmath130 of the unknown polygonal scatterer . a few detecting waves with the incident directions chosen among the set @xmath151 , @xmath152 ,
are sent off for locating the sides of the polygonal scatterer .
then the far - field data are measured and collected at 360 equidistant observation angles along the unit circle .
the far - field data generated on the unit circle are then subjected pointwise to certain uniform random noise .
the uniform random noise in magnitude as well as in direction is added according to the following formula , @xmath153 where @xmath154 and @xmath155 are two uniform random numbers , both ranging from -1 to 1 , and @xmath156 represents the noise level . for polygonal obstacles , we shall test sound - soft and sound - hard , noise - free and noisy cases , respectively . in the noise - free case , it is well - known that the far - field data is analytic and thus very smooth . as a consequence ,
the local maximum behavior of phaseless far - field data is clear from its polar graph as shown in fig .
[ fig:2 ] when there exists no noise .
while in the noisy case , we always add to the exact far - field data a uniform noise of @xmath157 and use it as the noisy measurement data , which is inevitable from the practical point of view .
but the local maximum behavior of phaseless far - field data might be corrupted by the ups and downs of random noise which cause fictitious and/or more local mamima than usual .
the cure out of the dilemma is to add a preprocessing step to filter the raw noisy data . in our tests ,
a fourier filtering stage is applied to the noisy far - field data in advance .
the filtered measurement data are thus used as in the recovery scheme , which gives us better reconstructions than using raw data by our experience .
plot of @xmath59 in polar coordinates corresponding to a sound - soft triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation angles @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] plot of @xmath59 in polar coordinates corresponding to a sound - soft triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation angles @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] ( a)(b ) plot of @xmath59 in polar coordinates corresponding to a sound - soft triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation angles @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] plot of @xmath59 in polar coordinates corresponding to a sound - soft triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation angles @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows . , title="fig:",scaledwidth=43.0% ] ( c)(d ) * example 1 . a triangle .
* the obstacle is chosen to be a triangle with three vertices displaced at @xmath163 , @xmath164 and @xmath165 , respectively . in this test
, we send off four detecting waves from north , east , south and east with @xmath166 , respectively .
firstly , we test a sound - soft triangular obstacle .
the initial position @xmath130 of the polygonal scatterer is detected by scheme i ( see @xcite ) and obtained by taking the position with maximum indicator function value .
the initial guess of the location point is found to be @xmath167 , denoted by a red star in fig .
[ fig : ex1:ss : reconstruction ] .
we plot the square power of the phaseless far - field data in fig . [
fig : ex1:ss : ff ] .
it can be seen from fig .
[ fig : ex1:ss : ff ] that the phaseless data display significant maxima along the forward - scattering ( or incident direction ) directions within the shadow region of the obstacle in all four plots . except the forward - scattering directions ,
we find several other directions with local maximum behavior in the polar plots and determine the three critical observation angles with the largest magnitude of phaseless data , which is indicated by the red arrows in fig .
[ fig : ex1:ss : ff](a ) , ( c ) and ( d ) , respectively . with the respective set of incident and critical observation directions obtained in hand , one can determine the three exterior unit normal directions of the sides of the polygonal obstacle .
the rest of the work is reduced to be a three - dimensional nonlinear least - squares minimization problem .
the final reconstruction results are shown in fig .
[ fig : ex1:ss : reconstruction ] , which is quite satisfactory in both noise - free and noisy cases .
it can be observed that we can obtain better reconstruction by using detecting waves with relatively high wave number for this sound - soft polygonal scatterer .
this can be explained in the following way .
the detecting wave with high frequency gives a more focused reflected beam and thus yields a better approximation of the physical optics .
the higher the frequency of the detecting wave , the more accurate determination of the outward normal directions and thus we can obtain better reconstruction plots .
moreover , we see that the proposed recovery scheme is tolerable to relatively high level of noise and performs robust in the noisy case .
reconstruction of the triangular sound - soft scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise and ( d ) @xmath168 without noise . , title="fig:",scaledwidth=48.0% ] reconstruction of the triangular sound - soft scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise and ( d ) @xmath168 without noise .
, title="fig:",scaledwidth=48.0% ] ( a)(b ) reconstruction of the triangular sound - soft scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise and ( d ) @xmath168 without noise . ,
title="fig:",scaledwidth=48.0% ] reconstruction of the triangular sound - soft scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise and ( d ) @xmath168 without noise . , title="fig:",scaledwidth=48.0% ] ( c)(d ) next , we keep the experimental settings unchanged except replacing the obstacle by a sound - hard triangular scatterer . as before
, the location point is detected to be @xmath169 using scheme i in @xcite , denoted by a red star in fig .
[ fig : ex1:sh : reconstruction ] . as shown in fig .
[ fig : ex1:sh : ff ] we plot the square power of the phaseless far - field data and indicate the first three critical observation angles within the backscattering aperture .
the final reconstruction results are shown in fig .
[ fig : ex1:sh : reconstruction ] , which performs as good as in the sound - soft case .
the effectiveness of the recovery scheme can be explained in a similar way as in the sound - soft case .
plot of @xmath59 in polar coordinates corresponding to a sound - hard triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] plot of @xmath59 in polar coordinates corresponding to a sound - hard triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows . ,
title="fig:",scaledwidth=43.0% ] ( a)(b ) plot of @xmath59 in polar coordinates corresponding to a sound - hard triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] plot of @xmath59 in polar coordinates corresponding to a sound - hard triangle due to an incident plane wave @xmath46 with @xmath158 and @xmath159 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 and @xmath162 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] ( c)(d ) reconstruction of the triangular sound - hard scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise.,title="fig:",scaledwidth=48.0% ] reconstruction of the triangular sound - hard scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise.,title="fig:",scaledwidth=48.0% ] ( a)(b ) reconstruction of the triangular sound - hard scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise.,title="fig:",scaledwidth=48.0% ] reconstruction of the triangular sound - hard scatterer with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise.,title="fig:",scaledwidth=48.0% ] ( c)(d ) * example 2
. a convex hexagon . * in the second example , the obstacle is chosen to be a sound - soft hexagon with six vertices displaced at @xmath170 , @xmath171 @xmath172 , @xmath173 @xmath174 and @xmath175 , respectively .
the location point is detected to be @xmath176 using scheme i in @xcite .
this example is much more challenging since there are multiple facets to be determined .
it is pointed out that the hexagonal obstacle has six sides and we only send off four detecting waves along incident directions @xmath177 . in this case , the unknown number of sides is larger than that of incident directions .
but from fig .
[ fig : ex2:ss : ff ] , one can reveal sufficient critical observation angles from those polar plots .
we identify six critical observation angles with relatively large local maximum values as indicated in red arrows in fig .
[ fig : ex2:ss : ff ] .
the final reconstruction results are obtained by solving a six - dimensional nonlinear problem and are shown in fig .
[ fig : ex2:ss : reconstruction ] .
it is again observed that the reconstruction performs better with higher frequency detecting waves . with the combining effect of increasing sides and noise , our recovery scheme still yield quite reasonable approximation to the original hexagon .
plot of @xmath59 in polar coordinates corresponding to a convex sound - soft hexagon due to an incident plane wave @xmath46 with @xmath158 and @xmath178 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 , @xmath162 , @xmath179 , @xmath180 and @xmath181 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] plot of @xmath59 in polar coordinates corresponding to a convex sound - soft hexagon due to an incident plane wave @xmath46 with @xmath158 and @xmath178 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 , @xmath162 , @xmath179 , @xmath180 and @xmath181 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] ( a)(b ) plot of @xmath59 in polar coordinates corresponding to a convex sound - soft hexagon due to an incident plane wave @xmath46 with @xmath158 and @xmath178 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 , @xmath162 , @xmath179 , @xmath180 and @xmath181 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] plot of @xmath59 in polar coordinates corresponding to a convex sound - soft hexagon due to an incident plane wave @xmath46 with @xmath158 and @xmath178 from ( a ) to ( d ) .
the selected critical observation directions @xmath160 , @xmath161 , @xmath162 , @xmath179 , @xmath180 and @xmath181 are highlighted by red arrows .
, title="fig:",scaledwidth=43.0% ] ( c)(d ) reconstruction of the sound - soft hexagonal obstacle with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise . , title="fig:",scaledwidth=48.0% ] reconstruction of the sound - soft hexagonal obstacle with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise .
, title="fig:",scaledwidth=48.0% ] ( a)(b ) reconstruction of the sound - soft hexagonal obstacle with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise . , title="fig:",scaledwidth=48.0% ] reconstruction of the sound - soft hexagonal obstacle with ( a ) @xmath158 with @xmath157 noise , ( b ) @xmath158 without noise , ( c ) @xmath168 with @xmath157 noise , and ( d ) @xmath168 without noise .
, title="fig:",scaledwidth=48.0% ] ( c)(d )
in this paper , we develop an inverse scattering scheme of recovering a sound - hard or sound - soft polyhedral obstacle by only a few backscattering far - field measurements .
it has been a very challenging issue in the literature on recovering an obstacle by minimum measurement data .
we believe that the results in this work make some important contribution to this challenging issue .
the proposed scheme proceeds with two steps .
first , one uses the local maximum behavior of the modulus of the backscattering far - field data to determine the exterior normal direction of each of the side / face of the obstacle .
then , one can solve a small - scale finite dimensional algebraic problem to completely recover the obstacle . in order to justify the local maximum behavior of the modulus of the far - field data
, we made essential use of the high - frequency asymptotics of the acoustic scattering .
our method can be extended to recovering non - convex obstacles , as well as to the electromagnetic scattering problems , which we shall report in the forthcoming work .
this work was supported by the nsf of china under the grant no .
11201453 and no.11371115 , and the frg and startup funds from the hong kong baptist university . | we propose an inverse scattering scheme of recovering a polyhedral obstacle in @xmath0 , @xmath1 , by only a few high - frequency acoustic backscattering measurements .
the obstacle could be sound - soft or sound - hard .
it is shown that the modulus of the far - field pattern in the backscattering aperture possesses a certain local maximum behavior , from which one can determine the exterior normal directions of the front sides / faces . then by using the phaseless backscattering data corresponding to a few incident plane waves with suitably chosen incident directions , one can determine the exterior unit normal vector of each side / face of the obstacle . after the determination of the exterior unit normals , the recovery is reduced to a finite - dimensional problem of determining a location point of the obstacle and the distance of each side
/ face away from the location point . for the latter reconstruction , we need make use of the far - field data with phases .
numerical experiments are also presented to illustrate the effectiveness of the proposed scheme . * keywords*. inverse scattering , polyhedral obstacle , backscattering , phaseless * mathematics subject classification ( 2010 ) * : 78a46 , 35q60 . |
the discovery of the top quark at fermilab fulfilled the three - family quark picture in the standard electroweak model .
up to now , some knowledge on the mass spectra of @xmath2 and @xmath3 quarks has been accumulated through both experimental and theoretical ( or phenomenological ) attempts @xcite .
the ratios of quark mass eigenvalues are obtainable after one renormalizes them to a common reference scale , e.g. , @xmath4 gev or @xmath5 .
there exists a clear mass hierarchy in each quark sector : @xmath6 in comparison , the masses of three charged leptons manifest a similar hierarchical pattern @xcite .
quark mass eigenstates are related to quark weak ( flavor ) eigenstates by the kobayashi - maskawa ( km ) matrix @xmath7 @xcite , which provides a quite natural description of flavor mixings and @xmath1 violation in the standard model . to date , many experimental constraints on the magnitudes of the km matrix elements have been achieved .
the unitarity of @xmath7 together with current data requires a unique hierarchy among the nine matrix elements @xcite : @xmath8 here @xmath9 is a necessary condition for the presence of @xmath1 violation in the km matrix . how to understand the hierarchies of quark masses and flavor mixings is an important but unsolved problem in particle physics .
a natural approach to the final solution of this problem is to look for the most favorable pattern of quark mass matrices ( see , e.g. , refs .
@xcite ) , which can account for all low - energy phenomena of quark mixings and @xmath1 violation .
the relevant symmetries hidden in such phenomenological schemes are possible to provide useful hints toward the dynamical details of fermion mass generation .
it has been speculated by some authors that the realistic fermion mass matrices could arise from the flavor permutation symmetry and its spontaneous or explicit breaking @xcite . under exact @xmath10 symmetry the mass spectrum for either up or down quark sector
consists of only two levels : one is of 2-fold degeneracy with vanishing mass eigenvalues , and the other is nondegenerate ( massive ) .
an appropriate breakdown of the above symmetry may lead to the observed mass hierarchy and flavor mixings .
although the way to introduce the minimum number of free parameters for permutation symmetry breaking is technically trivial , its consequences on quark mixings and @xmath1 violation may be physically instructive and may even shed some light on the proper relations between the km matrix elements and quark mass ratios .
indeed there has not been a satisfactory symmetry breaking pattern with enough predictive power in the literature . in this work
we first stress that some observed properties of the km matrix can be interpreted by the quark mass hierarchy without the assumption of specific mass matrices . in the quark mass limits such as @xmath11 , @xmath12 or @xmath13
, we find that simple but instructive relations between the km matrix elements and quark mass ratios are suggestible from current experimental data .
then we present a new quark mass _ ansatz _ through the explicit breakdown of flavor permutation symmetry at the weak scale ( @xmath14 gev ) .
this _ ansatz _ contains seven free parameters , thus it can give rise to three predictions for the phenomena of quark mixings and @xmath1 violation .
the typical results are @xmath15 , @xmath16 and @xmath17 in the leading order approximation .
prescribing the same _ ansatz _ at the supersymmetric grand unified theory ( gut ) scale ( @xmath18 gev ) , we derive the renormalized quark mass matrices at @xmath5 for small @xmath19 ( the ratio of higgs vacuum expectation values in the minimal supersymmetric model ) .
we also renormalize some relations between the km matrix elements and quark mass ratios at @xmath5 for arbitrary @xmath19 , and find that the relevant results are in good agreement with experimental data .
the scale - independent predictions of our _ ansatz _ for the characteristic measurables of @xmath1 asymmetries in weak @xmath20 decays , i.e. , @xmath21 , @xmath22 and @xmath23 , can be tested at the forthcoming kek and slac @xmath20-meson factories .
the remaining part of this paper is organized as follows .
some qualitative implications of the quark mass hierarchy on the km matrix elements , which are almost independent of the specific forms of quark mass matrices , are discussed in section 2 . in section 3
we suggest a new quark mass _ ansatz _ from the flavor permutation symmetry breaking at the weak scale , and study its various consequences on flavor mixings and @xmath1 violation .
the same _ ansatz _ is prescribed at the supersymmetric gut scale in section 4 . by use of the one - loop renormalization group equations , we run the mass matrices from @xmath24 to @xmath5 and then discuss the renormalized relations between the km matrix elements and quark mass ratios . section 5
is devoted to a brief summary of this work .
without loss of any generality , the up and down quark mass matrices ( denoted by @xmath25 and @xmath26 , respectively ) can be chosen to be hermitian .
after the diagonalization of @xmath25 and @xmath26 through the unitary transformations @xmath27 one obtains the km matrix @xmath28 , which describes quark flavor mixings in the charged current . explicitly , the km matrix elements read @xmath29 depending upon the quark mass ratios @xmath30 , @xmath31 ( from @xmath32 ) and @xmath33 , @xmath34 ( from @xmath35 ) as well as other parameters of @xmath25 and @xmath26 ( e.g. , the non - trivial phase shifts between @xmath25 and @xmath26 ) . in view of the distinctive mass hierarchy in eq .
( 1.1 ) , we find that some interesting properties of @xmath7 can be interpreted without the assumption of specific forms of @xmath25 and @xmath26 .
* a. @xmath36 and @xmath37 in the limits @xmath12 and @xmath13 * since the mass spectra of up and down quarks are absolutely dominated by @xmath38 and @xmath39 respectively , the limits @xmath12 and @xmath13 are expected to be very reliable when we discuss flavor mixings between @xmath40 and @xmath41 . in this case , the effective mass matrices turn out to be two @xmath42 matrices and the resultant flavor mixing matrix ( i.e. , the cabibbo matrix @xcite ) can not accommodate @xmath1 violation .
the magnitudes of @xmath43 and @xmath44 can be obtained from eq .
( 2.2 ) , since @xmath45 holds for both sectors in the above - mentioned mass limits .
we find that @xmath46 is a straightforward result guaranteed by the unitarity of @xmath32 and @xmath35 .
the current experimental data , together with unitary conditions of the @xmath47 km matrix , have implied @xcite @xmath48 which is insensitive to allowed errors of the wolfenstein parameters @xmath49 , @xmath50 and @xmath51 @xcite . from the discussions above we realize that the near equality of @xmath36 and @xmath37 is in fact a natural consequence of @xmath52 and @xmath53 .
the magnitude of @xmath43 ( or @xmath44 ) must be a function of the mass ratios @xmath30 and @xmath33 in the limits @xmath12 and @xmath13 , if @xmath25 and @xmath26 have parallel or quasi - parallel structures .
considering the experimentally allowed regions of @xmath30 ( @xmath54 @xcite ) , @xmath55 ( @xmath56 @xcite ) and @xmath36 ( @xmath57 @xcite ) , one may guess that @xmath36 is dominated by @xmath58 but receives small correction from @xmath59 .
indeed such an instructive result for @xmath36 or @xmath37 can be derived from @xmath42 hermitian mass matrices of the form @xcite @xmath60 where @xmath61 . denoting the phase difference between @xmath62 and @xmath63 as @xmath64
, we obtain @xmath65 although the @xmath42 flavor mixing matrix can not accommodate @xmath1 violation , the phase shift @xmath64 is non - trivial on the point that it sensitively determines the value of @xmath36 . for illustration , we calculate the allowed region of @xmath64 as a function of @xmath30 in fig .
it is clear that the possibilities @xmath66 , @xmath67 and @xmath68 have all been ruled out by current data on @xmath43 and @xmath55 , since @xmath69 is expected to be true .
we conclude that the presence of @xmath64 in the quark mass _ ansatz _ above is crucial for correct reproduction of @xmath36 and @xmath37 .
such a non - trivial phase shift will definitely lead to @xmath1 violation , when the limits @xmath12 and @xmath13 are discarded . * b. @xmath70 and @xmath71 in the limit @xmath11 * considering the fact that @xmath72 and @xmath73 are negligibly small in the mass spectra of up and down quarks
, one can take the reasonable limit @xmath11 to discuss flavor mixings between the second and third families . in this case
, there is no mixing between @xmath40 and @xmath41 or between @xmath40 and @xmath74 .
thus @xmath75 holds for both up and down mass matrices , and then we get @xmath76 .
the relation @xmath77 is straightforwardly obtainable from eq .
( 2.2 ) by use of the unitary conditions of @xmath32 and @xmath35 .
in contrast , the present data and unitarity of the km matrix requires @xcite @xmath78 we see that the near equality between @xmath70 and @xmath71 can be well understood , because the quark mass limit @xmath11 is a good approximation for @xmath25 and @xmath26 .
we expect that @xmath70 and @xmath71 are functions of the mass ratios @xmath31 and @xmath34 in the limit @xmath11 .
current experimental data give @xmath79 @xcite , while @xmath80 @xcite and @xmath81 @xcite are allowed .
thus @xmath70 ( or @xmath71 ) should be dominated by @xmath34 , and it may get a little correction from @xmath31 . to obtain a linear relation among @xmath82 , @xmath34 and @xmath31 in the leading order approximation , one can investigate mass matrices of the following hermitian form : @xmath83 where @xmath84 and @xmath85 for both quark sectors .
this generic pattern can also be regarded as a trivial generalization of the fritzsch _ ansatz _ , in which @xmath86 is assumed @xcite , but they have rather different consequences on the magnitudes of @xmath82 and @xmath87 . denoting @xmath88 , @xmath89 and @xmath90 , we find the approximate result @xmath91 one can see that @xmath92 holds approximately , if @xmath93 is comparable in magnitude with @xmath94 .
here the phase shift @xmath95 plays an insignificant ( negligible ) role in confronting eq .
( 2.8 ) with the experimental data on @xmath70 , since the term proportional to @xmath31 is significantly suppressed . to determine the values of @xmath93 and @xmath94
, however , one has to rely on a more specific _ ansatz _ of quark mass matrices .
* c. @xmath96 in @xmath13 and @xmath97 in @xmath12 * now let us take a look at the two smallest matrix elements of @xmath7 , @xmath98 and @xmath99 , in the quark mass limits .
taking @xmath13 , we have @xmath100 , because @xmath26 turns out to be an effective @xmath42 matrix in this limit . then the ratio of @xmath98 to @xmath70 reads @xmath101 obtained from eq .
contrary to common belief , @xmath96 is absolutely independent of the mass ratio @xmath33 in the limit @xmath102 !
therefore one expects that the left - handed side of eq .
( 2.9 ) is dominated by a simple function of the mass ratio @xmath30 , while the contribution from @xmath31 should be insignificant in most cases .
the present numerical knowledge of @xmath96 ( @xmath103 @xcite ) and @xmath30 ( @xmath54 @xcite ) implies that @xmath16 is likely to be true .
indeed such an approximate result can be reproduced from the fritzsch _ ansatz _ and a variety of its modified versions @xcite . in the mass limit @xmath12 , @xmath25 becomes an effective @xmath42 matrix , and then @xmath104 holds .
the ratio of @xmath99 to @xmath71 is obtainable from eq .
( 2.2 ) as follows : @xmath105 here again we find that @xmath97 is independent of both @xmath30 and @xmath31 in the limit @xmath12 , thus it may be a simple function of the mass ratios @xmath33 and @xmath34 .
the current data give @xmath106 @xcite , @xmath107 @xcite and @xmath108 @xcite .
we expect that @xmath109 has a large chance to be true in the leading order approximation .
note that this approximate relation can also be derived from the fritzsch _ ansatz _ or some of its revised versions @xcite .
the qualitative discussions above have shown that some properties of the km matrix @xmath7 can be well understood just from the quark mass hierarchy .
for example , @xmath110 and @xmath111 are natural consequences of arbitrary ( hermitian ) mass matrices with @xmath112 and @xmath113 respectively , where @xmath114 denote the mass eigenvalues of each quark sector . to a good degree of accuracy , @xmath36 and
@xmath37 are expected to be independent of the mass ratios @xmath31 and @xmath34 , while @xmath70 and @xmath71 are independent of @xmath30 and @xmath33 .
the ratios @xmath96 and @xmath97 may be simple functions of @xmath30 and @xmath33 , respectively , in the leading order approximations . these qualitative results should hold , in most cases and without fine tuning effects , for generic ( hermitian ) forms of @xmath25 and @xmath26 . they can be used as an enlightening clue for the construction of specific and predictive @xmath115 of quark mass matrices .
we are now in a position to consider the realistic @xmath47 mass matrices in no assumption of the quark mass limits .
such an _ ansatz _ should be able to yield the definite values of @xmath93 and @xmath94 in eq .
( 2.8 ) , and account for current experimental data on flavor mixings and @xmath1 violation at low - energy scales . * a. flavor permutation symmetry breaking * we start from the flavor permutation symmetry to construct quark mass matrices at the weak scale , so that the resultant km matrix can be directly confronted with the experimental data .
the mass matrix with the @xmath10 symmetry reads @xmath116 where @xmath117 denotes the mass eigenvalue of the third - family quark ( @xmath118 or @xmath119 ) .
note that @xmath120 is obtainable from another rank - one matrix @xmath121 through the unitary transformation @xmath122 , where @xmath123 to generate masses for the second- and first - family quarks , one has to break the permutation symmetry of @xmath120 to the @xmath124 and @xmath125 symmetries , respectively . here
we assume that the up and down mass matrices have the parallel symmetry breaking patterns , corresponding to the parallel dynamical details of quark mass generation .
we further assume that each symmetry breaking chain ( i.e. , @xmath126 or @xmath127 ) is induced by a single real parameter , and the possible phase shift between two quark sectors arises from an unknown dynamical mechanism . with the assumptions made above
, a new _ ansatz _ for the up and down mass matrices can be given as follows : @xmath128 \ ; , \ ] ] where @xmath129 and @xmath130 are real ( dimensionless ) perturbation parameters responsible for the breakdowns of @xmath131 and @xmath124 symmetries of @xmath120 , respectively .
in the basis of @xmath132 , the mass matrix @xmath133 takes the form @xmath134 which has three free parameters and three texture zeros .
diagonalizing @xmath135 through the unitary transformation @xmath136 , one can determine @xmath137 , @xmath129 and @xmath130 in terms of the quark mass eigenvalues . in the next - to - leading order approximations ,
we get @xmath138 then the elements of @xmath139 are expressible in terms of the mass ratios @xmath140 and @xmath141 .
the flavor mixing matrix can be given as @xmath142 , where @xmath143 is a diagonal phase matrix taking the form @xmath144 . here
@xmath145 denotes the phase shift between up and down mass matrices , and its presence is necessary for the _ ansatz _ to correctly reproduce both @xmath36 ( or @xmath37 ) and @xmath1 violation . *
b. flavor mixings and @xmath1 violation * calculating the km matrix elements @xmath36 and @xmath37 in the next - to - leading order approximation , we obtain @xmath146 this result is clearly consistent with that in eq .
the allowed region of @xmath64 has been shown by fig .
1 with the inputs of @xmath55 and @xmath36 .
we find @xmath147 for reasonable values of @xmath30 . in the leading order approximation of eq .
( 3.7 ) or eq . ( 2.5 ) , it is easy to check that @xmath37 , @xmath59 and @xmath58 form a triangle in the complex plane @xcite . in the next - to - leading order approximation , @xmath70 and @xmath71 can be given as @xmath148 \ ; . \ ] ] comparing between eqs .
( 3.8 ) and ( 2.8 ) , we get @xmath149 , determined by the quark mass _ ansatz _ in eq . ( 3.4 ) . by use of @xmath81 @xcite , we illustrate the allowed region of @xmath70 as a function of @xmath31 in fig .
2 , where the experimental constraint on @xmath70 ( @xmath150 @xcite ) has also been shown .
we see that the result of @xmath70 obtained in eq .
( 3.8 ) is rather favored by current data .
this implies that the pattern of permutation symmetry breaking ( i.e. , @xmath151 ) in eq .
( 3.4 ) may have a large chance to be true .
the ratios @xmath96 and @xmath97 are found to be @xmath152 to a good degree of accuracy . with @xmath153 .
the magnitude of @xmath154 may be as large as @xmath155 to @xmath156 for @xmath157 or @xmath68 , but it is only about @xmath158 for @xmath147 allowed by eq .
] by use of leutwyler s result @xmath107 @xcite , we get @xmath159 . in comparison , the current data together with unitarity of the @xmath47 km matrix yield @xmath160 @xcite .
the allowed region of @xmath96 is constrained by that of @xmath30 , which has not been reliably determined .
we find that @xmath161 is necessary for the quark mass _ ansatz _ in eq .
( 3.4 ) to accommodate the experimental result @xmath162 @xcite . in the leading order approximations , we have @xmath163 .
small corrections to these diagonal elements are obtainable with the help of the unitary conditions of @xmath7 .
if we rescale three sides of the unitarity triangle @xmath164 by @xmath165 , then the resultant triangle is approximately equivalent to that formed by @xmath44 , @xmath59 and @xmath58 in the complex plane @xcite .
this interesting result can be easily shown by use of eqs .
( 3.7 ) , ( 3.8 ) and ( 3.9 ) .
three inner angles of the unitarity triangle turn out to be @xmath166 in the approximations made above . at the forthcoming @xmath20-meson factories
, these three angles will be determined from @xmath1 asymmetries in a variety of weak @xmath20 decays ( e.g. , @xmath167 , @xmath168 and @xmath169 ) . for illustration ,
we calculate @xmath170 , @xmath171 and @xmath172 by use of eq .
( 3.10 ) and plot their allowed regions in fig .
3 . clearly
the quark mass _ ansatz _ under discussion favors @xmath173 , @xmath22 and @xmath23 .
these results do not involve large errors , and they can be confronted with the relevant experiments of @xmath20 decays and @xmath1 violation in the near future . finally we point out that @xmath1 violation in the km matrix , measured by the jarlskog parameter @xmath174 @xcite , can also be estimated in terms of quark mass ratios .
it is easy to obtain @xmath175 \sin\delta\phi \ ; . \ ] ] typically taking @xmath176 , @xmath177 , @xmath178 , @xmath179 and @xmath180 , we get @xmath181 .
this result is of course consistent with current data on @xmath1 violation in the @xmath182 mixing system @xcite .
it is interesting to speculate that the quark mass hierarchy and flavor mixings may arise from a certain symmetry breaking pattern in the context of supersymmetric guts @xcite . starting from the flavor permutation symmetry , here we prescribe the same _ ansatz _ for quark mass matrices as that proposed in eq .
( 3.4 ) at the supersymmetric gut scale @xmath24 .
for simplicity we use @xmath183 and @xmath184 , which correspond to @xmath185 in eq .
( 3.4 ) and @xmath135 in eq . ( 3.5 ) , to denote the mass matrices at @xmath24 in two different bases .
they are related to each other through the unitary transformation @xmath186 .
the flavor mixing matrix derived from @xmath183 ( or @xmath184 ) is denoted by @xmath187 .
the subsequent running effects of @xmath183 and @xmath187 from @xmath24 to @xmath5 can be calculated with the help of the renormalization group equations in the minimal supersymmetric standard model .
* a. renormalized mass matrices at @xmath5 * the simplicity of @xmath183 ( or @xmath184 ) may be spoiled after it evolves from @xmath24 to @xmath5 . to illustrate this point
, here we derive the renormalized mass matrices @xmath188 and @xmath189 at @xmath5 by use of the one - loop renormalization group equations for the yukawa matrices and gauge couplings @xcite . to get instructive analytical results
, we constrain the ratio of higgs vacuum expectation values @xmath19 to be small enough ( @xmath190 ) , so that all non - leading terms in the yukawa couplings different from that of the top quark can be safely neglected @xcite . in this approximation
, the evolution equations of @xmath188 and @xmath189 read @xmath191 \hat{m}^{\rm u}_0 \ ; , \nonumber \\
16 \pi^2 \frac{{\rm d } \hat{m}^{\rm d}_0}{{\rm d } \chi } & = & \left [ \frac{1}{v^2 } \left ( \hat{m}^{\rm u}_0 \hat{m}^{{\rm u}^{\dagger}}_0 \right ) - g_{\rm d } \right ] \hat{m}^{\rm d}_0 \ ; , \ ] ] where @xmath192 , @xmath193 and @xmath194 are functions of the gauge couplings @xmath195 ( @xmath196 ) , and @xmath197 is the overall higgs vacuum expectation value normalized to 175 gev .
for the charged lepton mass matrix @xmath198 , its evolution equation is dominated only by a linear term @xmath199 in the case of small @xmath19 .
thus the hermitian structure of @xmath198 will be unchanged through the running from @xmath24 to @xmath5 ( in our discussions the neutrinos are assumed to be massless ) . the quantity @xmath200 ( n = u , d or e ) obeys the following equation : @xmath201 \ ; , \ ] ] where @xmath202 and @xmath203 are coefficients in the context of the minimal supersymmetric standard model .
the values of @xmath204 , @xmath202 and @xmath203 are listed in table 1 .
c|ccccc @xmath205 & @xmath206 & @xmath207 & @xmath208 & @xmath203 & @xmath209 + + 1 & 13/9 & 7/9 & 3 & 11 & 0.127 + + 2 & 3 & 3 & 3 & 1 & 0.42 + + 3 & 16/3 & 16/3 & 0 & @xmath2103 & 1.44 + + in order to solve eq .
( 4.1 ) , we diagonalize @xmath188 through the unitary transformation @xmath211 . making the same transformation for @xmath189 , i.e. , @xmath212 , we obtain the simplified evolution equations as follows : @xmath213 \hat{m}^{{\rm u}^{\prime}}_0 \ ; , \nonumber \\
16 \pi^2 \frac{{\rm d}\hat{m}^{{\rm d}^{\prime}}_0}{{\rm d } \chi } & = & \left [ f^2_t \left ( \matrix { 0 & 0 & 0 \cr 0 & 0 & 0 \cr 0 & 0 & 1 } \right ) - g_{\rm u } \left ( \matrix { 1 & 0 & 0 \cr 0 & 1 & 0 \cr 0 & 0 & 1 } \right ) \right ] \hat{m}^{{\rm d}^{\prime}}_0 \ ; , \ ] ] where @xmath214 is the top quark yukawa coupling eigenvalue .
for simplicity in presenting the results , we define @xmath215 \ ; , \nonumber \\ \xi_i & = & \exp \left [ - \frac{1}{16\pi^2 } \int^{\ln ( m_x / m_z)}_0 f^2_i ( \chi ) ~ { \rm d}\chi \right ] \ ] ] with @xmath216 ( or @xmath217 ) . by use of eq .
( 4.2 ) and the inputs listed in table 1 , one can explicitly calculate the magnitude of @xmath218 .
we find @xmath219 , @xmath220 and @xmath221 for @xmath222 gev and @xmath223 gev .
the size of @xmath224 depends upon the value of @xmath19 and will be estimated in the next subsection .
solving eq .
( 4.3 ) and transforming @xmath225 back to @xmath226 , we get @xmath227 in the leading order approximation .
since @xmath228 can be easily determined from @xmath229 and @xmath230 in the approximation of @xmath231 made above , we explicitly express @xmath232 and @xmath233 as follows : @xmath234 \ ; \ ] ] with @xmath235 , @xmath236 and @xmath237 ; and @xmath238 \ ; \ ] ] with @xmath239 , @xmath240 , @xmath241 and @xmath242 if one takes @xmath243 , which leads to @xmath244 , @xmath245 and in turn @xmath246 , @xmath247 , @xmath248 and @xmath249 , then eqs .
( 4.6 ) and ( 4.7 ) recover the form of @xmath250 as assumed in eq .
( 3.4 ) . to a good degree of accuracy
, @xmath232 remains hermitian .
the hermiticity of @xmath233 is violated by @xmath251 , which would vanish if the top and bottom quark masses were identical ( i.e. , @xmath252 ) .
the presence of nonvanishing @xmath251 reflects the fact that @xmath38 dominates the mass spectra of both quark sectors @xcite .
of course , one can transform the mass matrices obtained in eqs .
( 4.6 ) and ( 4.7 ) into the basis of @xmath184 . in doing so
, we will find the inequality between ( 2,3 ) and ( 3,2 ) elements of @xmath253 , arising from @xmath251 .
* b. renormalized flavor mixings at @xmath5 * calculating the magnitudes of flavor mixings from @xmath183 or @xmath184 at @xmath24 , we can obtain the same asymptotic relations between the km matrix elements and quark mass ratios as those given in eqs .
( 3.7 ) , ( 3.8 ) and ( 3.9 ) .
now we renormalize such relations at the weak scale @xmath5 by means of the renormalization group equations .
the quantities @xmath224 and @xmath254 defined in eq .
( 4.4 ) will be evaluated below for arbitrary @xmath19 , so that one can get some quantitative feeling about the running effects of quark mass matrices and flavor mixings from @xmath24 to @xmath5 .
the one - loop renormalization group equations for quark mass ratios and elements of the km matrix @xmath187 have been explicitly presented by babu and shafi in ref .
@xcite . in view of the hierarchy of yukawa couplings and
quark mixing angles , one can make reliable analytical approximations for the relevant evolution equations by keeping only the leading terms .
it has been found that ( 1 ) the running effects of @xmath30 and @xmath33 are negligibly small ; ( 2 ) the diagonal elements of the km matrix have negligible evolutions with energy ; ( 3 ) the evolutions of @xmath255 and @xmath256 involve the second - family yukawa couplings and thus they are negligible ; ( 4 ) the km matrix elements @xmath257 , @xmath258 , @xmath259 and @xmath260 have identical running behaviors . considering these points as well as
the dominance of the third - family yukawa couplings ( i.e. , @xmath261 and @xmath262 ) , we get three key evolution equations in the minimal supersymmetric standard model : @xmath263 with @xmath264 , @xmath265 , @xmath266 or @xmath267 . in the same approximations ,
the renormalization group equations for the yukawa coupling eigenvalues @xmath261 , @xmath262 and @xmath268 read @xcite : @xmath269 where the quantities @xmath200 have been given in eq .
( 4.2 ) . with the typical inputs @xmath270 gev , @xmath271 gev , @xmath272 gev and those listed in table 1 , we calculate @xmath224 and @xmath254 for arbitrary @xmath19 by use of the above equations .
our result is illustrated in fig .
we see that @xmath273 for @xmath274 .
this justifies our approximation made previously in deriving @xmath232 and @xmath233 . within the perturbatively allowed region of @xmath19 @xcite
, @xmath254 may be comparable in magnitude with @xmath224 when @xmath275 .
in this case , the evolution effects of quark mass matrices and flavor mixings are sensitive to both @xmath261 and @xmath262 . clearly the analytical results of @xmath255 , @xmath256 , @xmath276 and @xmath277 as those given in eqs .
( 3.7 ) and ( 3.9 ) are almost scale - independent , i.e. , they hold at both @xmath278 and @xmath279 .
non - negligible running effects can only appear in the expression of @xmath258 or @xmath260 , which is a function of the mass ratios @xmath34 and @xmath31 ( see eq .
( 3.8 ) for illustration ) . with the help of eq . (
4.9 ) , we find the renormalized relation between @xmath258 ( or @xmath260 ) and the quark mass ratios at the weak scale @xmath5 : @xmath280 \ ; .
\ ] ] this result will recover that in eq .
( 3.8 ) if one takes @xmath243 ( i.e. , @xmath281 ) . using @xmath282 @xcite and taking @xmath283
typically , we confront eq .
( 4.11 ) with the experimental data on @xmath284 ( i.e. , @xmath285 @xcite ) . as shown in fig .
5 , our result is in good agreement with experiments for @xmath286 .
this implies that the quark mass pattern @xmath183 or @xmath184 , proposed at the supersymmetric gut scale @xmath24 , may have a large chance to survive for reasonable values of @xmath19 .
note that evolution of the @xmath1-violating parameter @xmath174 is dominated by that of @xmath287 .
note also that three sides of the unitarity triangle @xmath288 have identical running effects from @xmath24 to @xmath5 , hence its three inner angles are scale - independent and take the same values as those given in eq .
( 3.10 ) or fig .
3 . as a result
, measurements of @xmath289 , @xmath290 and @xmath291 in the forthcoming experiments of @xmath20 physics may check both the quark mass _ ansatz _ at the weak scale and that at the supersymmetric gut scale .
without the assumption of specific mass matrices , we have pointed out that part of the observed properties of flavor mixings can be well understood just from the quark mass hierarchy . in the quark mass limits such as @xmath11 , @xmath12 or @xmath13 , a few instructive relations between the km matrix elements and quark mass ratios are suggestible from current experimental data .
we stress that such _
ansatz_-independent results may serve as a useful guide in constructing the specific quark mass matrices at either low - energy scales or superheavy scales .
starting from the flavor permutation symmetry and assuming an explicit pattern of symmetry breaking , we have proposed a new quark mass _ ansatz _ at the weak scale .
we find that all experimental data on quark mixings and @xmath1 violation can be accounted for by our _ ansatz_. in particular , we obtain an instructive relation among @xmath70 , @xmath34 and @xmath31 in the next - to - leading approximation ( see eq . ( 3.8 ) ) .
the scale - independent predictions of our quark mass pattern , such as @xmath21 , @xmath22 and @xmath23 , can be confronted with the forthcoming experiments at kek and slac @xmath20-meson factories . with the same _ ansatz _ prescribed at the supersymmetric gut scale @xmath24 , we have derived the renormalized quark mass matrices at the weak scale @xmath5 for small @xmath19 and calculated the renormalized flavor mixing matrix elements at @xmath5 for arbitrary @xmath19 . except @xmath258 and @xmath260 ,
the other asymptotic relations between the km matrix elements and quark mass ratios are almost scale - independent .
we find that the renormalized result of @xmath258 ( or @xmath260 ) is in good agreement with the relevant experimental data for reasonable values of @xmath19 .
in this work we neither assumed a specific form for the charged lepton mass matrix nor supposed its relation with the down quark mass matrix within the supersymmetric gut framework .
of course , this can be done by following the strategy proposed in ref . @xcite .
then one may obtain the relations between @xmath73 , @xmath292 , @xmath39 and @xmath293 , @xmath294 , @xmath295 .
such an _ ansatz _ , based on the specific gut scheme and flavor permutation symmetry breaking , will be discussed somewhere else .
the author would like to thank a.i .
sanda for his warm hospitality and the japan society for the promotion of science for its financial support .
he is also grateful to h. fritzsch , a.i .
sanda and k. yamawaki for their useful comments on the topic of permutation symmetry breaking and on part of this work .
xing , nucl .
b ( proc . suppl . ) * 50 * , 24 ( 1996 ) ; nuovo cimento a * 109 * , 115 ( 1996 ) ; in the proceedings of the international europhysics conference on high energy physics , brussels , belgium , 1995 , edited by j. lemonne , c. vander velde , and f. verbeure ( world scientific , singapore , 1996 ) , p. 181 .
h. harari , h. haut , and j. weyers , phys .
b * 78 * , 459 ( 1978 ) ; y. koide , phys . rev .
d * 28 * , 252 ( 1983 ) ; phys .
d * 39 * , 1391 ( 1989 ) ; z. phys .
c * 45 * , 39 ( 1989 ) ; h. fritzsch , in the proceedings of the europhysics topical conference on flavor mixing in weak interactions , erice , italy , 1984 , edited by l.l .
chau ( plenum , new york , 1984 ) , p. 717 ; c. jarlskog , in the proceedings of the international symposium on production and decay of heavy hadrons , heidelberg , germany , 1986 , edited by k.r .
schubert and r. waldi ( desy , hamburg , 1986 ) , p. 331 ; p. kaus and s. meshkov , ann .
n. y. acad .
sci . * 578 * , 353 ( 1989 ) ; phys .
d * 42 * , 1863 ( 1990 ) ; m. tanimoto , phys .
d * 41 * , 1586 ( 1990 ) ; g.c .
branco , j.i .
silva - marcos , and m.n .
rebelo , phys .
b * 237 * , 446 ( 1990 ) ; h. fritzsch and j. plankl , phys .
b * 237 * , 451 ( 1990 ) ; c.e .
lee , c. lin , and y.w .
yang , phys .
d * 42 * , 2355 ( 1990 ) ; p.f .
harrison and w.g .
scott , phys .
b * 333 * , 471 ( 1994 ) ; g.c .
branco and j.i .
silva - marcos , phys .
b * 359 * , 166 ( 1995 ) . for a brief review of other relevant references ,
see : s. meshkov , in the proceedings of the global foundation international conference on unified symmetry in the small and in the large , coral gables , florida , 1993 , edited by b.n .
kursunoglu and a. perlmutter ( nova sci .
publ . , new york , 1994 ) , p. 195
see , e.g. , l.j . hall and a. rasin , phys .
b * 315 * , 164 ( 1993 ) ; d. du and z.z .
xing , phys .
d * 48 * , 2349 ( 1993 ) ; k. harayama , n. okamura , a.i .
sanda , and z.z .
xing , report no .
dpnu-96 - 36 ( 1996 ) .
s. dimopoulos , l.j .
hall , and s. raby , phys .
lett . * 68 * , 1984 ( 1992 ) ; g. anderson _
et al . _ ,
d * 47 * , 3702 ( 1993 ) ; p.
ramond , r.g .
roberts , and g.g .
ross , nucl .
b * 406 * , 19 ( 1993 ) ; a. kusenko and r. shrock , phys .
d * 49 * , 4962 ( 1994 ) ; m. leurer , y. nir , and n. seiberg , nucl .
b * 420 * , 468 ( 1994 ) ; l.e .
ibanez and g.g .
ross , phys .
b * 332 * , 100 ( 1994 ) ; k.s .
babu and r.n .
mohapatra , phys .
* 74 * , 2418 ( 1995 ) ; k.s .
babu and s.b .
barr , phys .
lett . * 75 * , 2088 ( 1995 ) ; l.j . hall and s.
raby , phys .
d * 51 * , 6524 ( 1995 ) ; k.c .
chou and y.l .
wu , phys .
d * 53 * , r3492 ( 1996 ) .
for a brief review with more extensive references , see : c.d .
froggatt , report no
. gutpa/96/02/1 ( to be published in the proceedings of the fifth hellenic school and workshops on elementary particle physics , corfu , september , 1995 ) . | we stress that the observed pattern of flavor mixings can be partly interpreted by the quark mass hierarchy without the assumption of specific quark mass matrices .
the quantitatively proper relations between the kobayashi - maskawa matrix elements and quark mass ratios , such as @xmath0 \ ; , \ ] ] are obtainable from a simple _ ansatz _ of flavor permutation symmetry breaking at the weak scale .
we prescribe the same _ ansatz _ at the supersymmetric grand unified theory scale , and find that its all low - energy consequences on flavor mixings and @xmath1 violation are in good agreement with current experimental data .
= 17.1 cm = 24.71 cm = -15.mm = -11 mm plus 1pt minus 1pt * dpnu-96 - 39 + august 1996 * * implications of the quark mass hierarchy on flavor mixings * zhi - zhong xing _ department of physics , nagoya university , chikusa - ku , nagoya 464 - 01 , japan _
pacs number(s ) : 12.15.ff , 11.30.hv , 11.30.pb , 12.10.dm |
in collision - poor plasmas from space large deviations from thermodynamic equilibrium can not be relaxed by the particle - particle ( coulomb ) collisions , but can presumably be constrained by the resulting kinetic instabilities .
thus , if the solar wind expands adiabatically the cgl invariants conserve @xcite leading to an indefinite increase of temperature in the direction parallel to the inteplanetary magnetic field , i.e. , @xmath0 .
however , the in - situ measurements do not confirm such an increase of their parallel temperature with heliocentric distance , but indicate bounds of the temperature anisotropy of plasma particles @xcite . because collisions are not efficient , the most invoked mechanism that can limit the increase of parallel temperature is the firehose instability @xcite .
the firehose instability driven by the anisotropic electrons with @xmath1 , also known as the electron firehose instability ( efhi ) , is particularly important as it can mediate a resonant transfer of ( free ) energy from electrons to protons @xcite .
this energy transfer from small to large scales is facilitated by the quasi - parallel efh modes , which are left - handed ( lh ) circularly polarized and have characteristic frequencies and growth rates in the range of the proton cyclotron frequency .
besides the propagating ( non - zero frequency ) modes predominant at small angles ( quasi - parallel ) with respect to the magnetic field direction , the firehose instability may destabilize an additional aperiodic ( non - propagating ) branch which exists only for oblique directions @xcite .
although it is well known that the suprathermal populations are ubiquitous in the solar wind @xcite , the anisotropic temperature is in general quantified by a bi - maxwellian distribution function , which is relevant only for the thermal core of the solar wind electrons . in this case the aperiodic fhi is found to grow faster than the propagating modes , and the instability thresholds approach well enough the limits of the core anisotropy reported by the observations @xcite . for anisotropies exceeding these thresholds ,
the free energy is dissipated by the resulting instabilities , which may also scatter particles back towards quasi - equilibrium states and prevent the anisotropy to grow @xcite . instead , for the suprathermal electrons from the solar wind the limits of their anisotropy
are markedly departed from the instability thresholds derived for bi - maxwellian populations , see figure 6 and the analysis in @xcite .
this disagreement may simply be motivated by the fact that suprathermal populations can not be properly described by the maxwellian distribution functions , but they can be accurately reproduced by the kappa power - laws @xcite .
ubiquitous in the solar wind and subsequent planetary environments , e.g. , terrestrial magnetosphere , see the review by @xcite , suprathermal electrons are more dilute but hotter than the core populations .
the relaxation through the particle - particle collisions is even less efficient in this case , but kinetic instabilities are expected to explain the limits of temperature anisotropy reported by the observations . in the present paper
we propose a refined analysis of the suprathermal electrons by using a bi - kappa distribution function to describe the anisotropy of these populations . in the limit of a high power - index @xmath2
the ( bi-)kappa distribution function reduces to a ( bi-)maxwellian . @xcite and @xcite
have used the bi - kappa model to quantify the velocity distributions and the principal properties of the suprathermal electrons in the solar wind , e.g. , the components of the anisotropic temperature , parallel ( @xmath3 ) and perpedicular ( @xmath4 ) to the magnetic field direction .
the suprathermal electrons are found to be highly anisotropic and with a predominant excess of parallel temperature susceptible to the fhi .
the bi - kappa model was also extensively invoked in theories of dispersion and stability by adopting two alternative assumptions for the temperature of kappa populations , to be either dependent or independent of the power - index @xmath5 .
studies of the fhi @xcite assume @xmath5-independent temperatures , and find , contrary to the expectations , that the instability is inhibited by the suprathermals and the instability thresholds do not approach but depart even more from the anisotropy bounds of the solar wind suprathermal electrons .
however , from a recent analysis on the applicability of kappa distributions @xcite it becomes evident that a representation with a @xmath5-dependent temperature may provide a more natural interpretation of the suprathermal populations for three fundamental reasons : ( 1 ) it corresponds to a maxwellian limit which reproduces more accurately the thermal ( core ) population enabling for a direct and realistic comparison @xcite ; ( 2 ) the kinetic instabilities show a systematic stimulation in the presence of suprathermal electrons @xcite as one may expect from the excess of free energy acumulated by these populations ; and ( 3 ) the observations show strong evidence of @xmath5-dependent temperatures , which increase in the presence of suprathermal populations , i.e. , temperatures increase with decreasing the power - index @xmath5 @xcite . motivated by these premises , here we re - analyse the instability of the efh mode by modeling the suprathermal electrons with a bi - kappa approach with @xmath5-dependent temperatures . in this preliminary analysis
we restrict to the same parallel ( non - zero frequency ) modes studied before by @xcite .
the bi - kappa approach is introduced in section 2 , enabling us to derive the dispersion relation for the fhi modes .
the main features of the instability , are derived and discussed in section 3 .
in addition , the efh thresholds are compared with the observations of the suprathermal electron anisotropy .
the results of the present work are summarized in section 4 .
we first introduce the analytical model for the velocity distributions of suprathermal electrons detected in space plasmas @xcite . the suprathermal ( halo )
electrons are assumed to be a gyrotropic component ( isotropic in the plane transverse to the magnetic field ) with a bi - axis temperature anisotropy @xmath6 , where @xmath7 and @xmath8 denote directions relative to the magnetic field .
the distribution of suprathermal electrons in velocity space with polar coordinates @xmath9 is described by a bi - kappa distribution function @xmath10 \over \kappa^{1/2 } \gamma[\kappa - 1/2]}\notag \\ & \times \left(1 + { v_{\parallel}^2\over \kappa \theta_{\parallel}^2 } + { v_{\perp}^2\over \kappa \theta_{\perp}^2 } \right)^{-\kappa-1 } , \label{e1}\end{aligned}\ ] ] which is normalized to unity , and where @xmath11 are thermal velocities defined by , respectively , the parallel and perpendicular temperatures as moments of second order @xmath12 @xmath13 the bi - kappa simply reduces to a bi - maxellian in the limit of a very large @xmath2 @xmath14 with @xmath15 @xmath16 notice in case that the temperature of suprathermal electrons decreases with increasing the power - index @xmath5 and reaches a minimum for the maxwellian limit . in the direction parallel to the magnetic field ( @xmath17 ) ,
the electromagnetic ( em ) modes are decoupled from the electrostatic oscillations , and are described by the following general dispersion relation @xcite @xmath18 , \label{e7}\end{aligned}\ ] ] where @xmath19 and @xmath20 are respectively , the frequency and the wavenumber of the plasma modes , @xmath21 is the speed of light in vacuum , @xmath22 is the gyrofrequency for the particles of sort @xmath23 , e.g. , @xmath24 for electrons and @xmath25 for protons , respectively , and `` @xmath26 '' describes the circularly polarized em modes with right - hand ( rh ) and left - hand ( lh ) polarizations , respectively . for the advanced model introduced in equation ( [ e1 ] )
the dispersion relation becomes @xmath27 , \label{e8}\end{aligned}\ ] ] where @xmath28 is the temperature anisotropy , @xmath29 is the kappa plasma dispersion function @xcite of argument @xmath30 in the maxwellian limit this function reduces to the standard plasma dispersion function @xcite @xmath31 of argument @xmath32 note that for our model introduced in eqs .
( [ e1])([e6 ] ) , the anisotropy does not depend on @xmath5 , i.e. , @xmath33 .
we investigate the efhi , which is a lh em mode driven unstable by an excess of electron temperature in parallel direction @xmath34 , i.e. , @xmath35 . according to ( [ e8 ] )
, the dispersion relation describing these modes can be rewritten with normalized quantities as follows @xmath36 \notag \\
+ \frac{\tilde{\omega}}{\tilde{k } \sqrt{\beta _ { e,\parallel}^m /\theta } } z\left ( \frac{\tilde{\omega}-1 } { \tilde{k}\sqrt{\beta_{e,\parallel}^m / \theta}}\right ) = \tilde{k}^{2 } , \label{e13}\end{aligned}\ ] ] where protons are assumed maxwellian and isotropic @xmath37 , and @xmath38 , @xmath39 , @xmath40 is the proton / electron mass ratio , @xmath41 @xmath42 is the electron/ proton parallel temperature ratio in the maxwellian limit for both species , and @xmath43 is the parallel electron beta parameter in the maxwellian limit @xmath2 .
the dispersion relation for bi - maxwellian distributed electrons can be obtained from eq .
( [ e13 ] ) only by changing @xmath44 with the maxwellian plasma dispersion function from ( [ e11 ] ) .
we have solved the dispersion relation ( [ e13 ] ) numerically , and analyzed the unstable firehose solutions . in this section
we present the main features of the efhi , namely , growth rates , wave - frequencies and wave - numbers , as well as the anisotropy thresholds , and restrict our analysis only to the unstable regimes controlled mainly by the suprathermal electrons .
the effects of suprathermal populations are triggered by their temperature anisotropy and their abundance , which is quantified by the finite ( especially low ) values of the power - index @xmath5 .
firstly , we examine the growth rates and the wave - frequency of the efh instability for different plasma regimes conditioned in principal by the ( parallel ) plasma beta parameter , @xmath45 , the electron anisotropy @xmath46 , and the power - index @xmath5 .
the regimes identified in figures [ f1 ] and [ f2 ] are specific to the firehose instability , when a magnetized plasma becomes penetrable by the lh electromagnetic fluctuations propagating parallel to the magnetic field with frequencies higher than the proton cyclotron frequency .
all the unstable modes , i.e. , with @xmath47 in figure [ f1 ] , exhibit this property that becomes evident in figure [ f2 ] , where their wave - number dispersion extends to high frequencies exceeding @xmath48 . in the presence of suprathermals ,
i.e. , at low values of @xmath5 , the range of unstable wave - nunmbers is restrained , but the wave - frequencies and the instability growth - rates are enhanced .
these effects are in general stimulated by increasing the plasma beta parameter @xmath49 , the temperature anisotropy and the electron - proton temperature contrast @xmath50
. plots evidencing the influence of @xmath50 are not shown here , but details about this influence are explicitly given in the text .
the unstable solutions displayed in figures [ f1]-[f4 ] are obtained for the same value of this parameter , namely , for @xmath51 in accordance to the observations in the slow solar wind @xcite . 2 , 2.3 , 3 , 6 , @xmath52 , on the growth rates of the efh instability for different plasma beta explicitly given in each panel.,title="fig:",width=302 ] 2 , 2.3 , 3 , 6 , @xmath52 , on the growth rates of the efh instability for different plasma beta explicitly given in each panel.,title="fig:",width=302 ] 2 , 2.3 , 3 , 6 , @xmath52 , on the growth rates of the efh instability for different plasma beta explicitly given in each panel.,title="fig:",width=302 ] 2 , 2.3 , 3 , 6 , @xmath52 ) on the wave - frequency of the efh instability for the same cases considered in fig . [ f1].,title="fig:",width=302 ] 2 , 2.3 , 3 , 6 , @xmath52 ) on the wave - frequency of the efh instability for the same cases considered in fig . [ f1].,title="fig:",width=302 ] 2 , 2.3 , 3 , 6 , @xmath52 ) on the wave - frequency of the efh instability for the same cases considered in fig
. [ f1].,title="fig:",width=302 ] at higher values of @xmath5 the instability conditions may be not satisfied and the electromagnetic modes are damped , e.g. , @xmath53 for @xmath54 in figure [ f1 ] , middle and bottom panels . for these modes , the wave - frequency dispersion curves displayed in figure [ f2 ] have a different allure , showing an asymptotic increase similar to the ion ( proton ) cyclotron modes with frequencies always smaller than @xmath48 .
these are lh modes damped by the protons and therefore limited only to the large ( proton ) scales . at smaller scales controlled by the electrons ( higher wave - numbers ) these modes
change ( mode conversion ) to rh polarization ( i.e. , the wave - frequency displayed in figure [ f2 ] becomes negative ) which is more specific to the electron whistlers . in figures [ f3 ] and [ f4 ]
we show that these lh - polarized modes with a wave - number dispersion resembling that of the electromagnetic ion cyclotron ( emic ) modes can be destabilized by the anisotropic bi - kappa distributed electrons , see middle and bottom panels .
this is a new regime of the efhi destabilizing only the low - frequency branch of the lh modes with wave - frequency showing an asymptotic increase of their wave - frequencies but remaining always below @xmath48 . to establish this regime the kinetic effects of the electrons
are also tempered by considering lower values of plasma beta , and the instability is triggered only by the anisotropic distributions with sufficiently low @xmath5 , e.g. , @xmath55 in figures [ f3 ] and [ f4 ] .
furthermore , in this case , both the ( maximum ) growth - rates and the range of the unstable wave - numbers are considerably enhanced by increasing the presence of suprathermals , i.e. , lowering the values of @xmath5 .
again , these features seems to be more specific to the instability of the cyclotron modes @xcite .
the transition between the classical efh solutions ( exemplified in figures [ f1 ] and [ f2 ] ) and the new regime of a low - frequency efhi is suggestively shown by the top panels in figures [ f3 ] and [ f4 ] . in these panels
we have unstable solutions specific to both these regimes : the solid - line solution obtained for @xmath56 is a classical firehose , while the next long - dashed - line solution obtained for @xmath57 is already more specific to the new regime of efhi . in this case
it is only the power - index @xmath5 that may switch between these two regimes , but a direct comparison of the other plasma parameters in figures [ f1]-[f4 ] , clearly shows that these regimes are also conditioned by the temperature anisotropy , the plasma beta , and the temperature contrast between electrons and protons . 2 , 2.3 , 3 , 6 , @xmath52 , on the growth rates of the efh instability for a lower @xmath58.,title="fig:",width=321 ] 2 , 2.3 , 3 , 6 , @xmath52 , on the growth rates of the efh instability for a lower @xmath58.,title="fig:",width=321 ] 2 , 2.3 , 3 , 6 , @xmath52 , on the growth rates of the efh instability for a lower @xmath58.,title="fig:",width=321 ] 2 , 2.3 , 3 , 6 , @xmath52 ) on the wave - frequency of the efh instability for the same cases considered in fig . [ f3].,title="fig:",width=302 ] 2 , 2.3 , 3 , 6 , @xmath52 ) on the wave - frequency of the efh instability for the same cases considered in fig . [ f3].,title="fig:",width=302 ] 2 , 2.3 , 3 , 6 , @xmath52 ) on the wave - frequency of the efh instability for the same cases considered in fig .
[ f3].,title="fig:",width=302 ] ) for maximum growth rates @xmath59 with the temperature anisotropy measured in the solar wind which is displayed using a scatter plot data in the top panel and a histogram data in the bottom panel.,title="fig:",width=302 ] + ) for maximum growth rates @xmath59 with the temperature anisotropy measured in the solar wind which is displayed using a scatter plot data in the top panel and a histogram data in the bottom panel.,title="fig:",width=302 ] in the second part of this section we analyze the anisotropy thresholds of the instability .
these thresholds represent plasma conditions associated with given values of the maximum growth - rate , usually small values , e.g. , @xmath60 , @xmath61 , approaching the marginal stability @xmath62 . in figure
[ f5 ] we display the instability thresholds associated with @xmath63 and derived for different values of the electron power - index @xmath5 .
these are isocontours of the electron temperature anisotropy @xmath46 as a function of the parallel electron plasma beta @xmath64 , fitted to an inverse correlation law of the form @xcite @xmath65 the values obtained for the fitting parameters @xmath23 and @xmath66 can be found in table [ t1 ] . for the plasma beta parameter
we consider an extended range of values @xmath67 relevant for the electron halo populations in the solar wind @xcite .
higher values of @xmath64 , associated with hotter plasmas or less intense magnetic fields , means lower deviations from isotropy to trigger the instability .
the effects of suprathermal electrons is reconfirmed here by a systematic stimulation of the ( maximum ) growth - rates with decreasing @xmath5 . as a consequence ,
the anisotropy thresholds are found to be markedly lowered in the presence of suprathermals , and this effect may be enhanced by increasing the temperature contrast between electrons and protons .
larger variations of the anisotropy thresholds are obtained at lower values of @xmath5 .
the instability thresholds are compared in figure [ f5 ] with the observational data of the electron halo populations in the slow solar wind ( @xmath68 km / s ) , which are displayed in the top panel as a scatter plot , and in the bottom panel as a histogram counting the number of events within a color logarithmic scale .
this data set comprises more than 120 000 events detected by three space missions ( helios 1 , cluster ii , and ulysses ) at different heliocentric distances ( in the interval 0.33.95 au ) in the ecliptic . the details about the electron analyzers used by these missions , and the methods of correction and reconstruction of the 3d velocity distribution functions
can be found in @xcite .
these authors have used the same set of events to analyze the temperature anisotropy of the main electron populations , namely , the thermal core and suprathermal halo , and the most plausible constraints exercised on their temperature anisotropy by different physical mechanisms , e.g. , collisions and kinetic instabilities .
however , @xcite have limited to investigate in detail only the bi - maxwellian core anisotropy finding that the particle - particle collisions still may have an effect to constrain low levels of anisotropy , while the kinetic instabilities occur for larger deviations from isotropy , which exceed their thresholds .
indeed , the instability thresholds predicted by a bi - maxwellian model are found to shape very well the limits of the core anisotropy , but the same thresholds can not explain the limits observed for the temperature anisotropy of suprathermal electrons . figure 6 from @xcite presents such a comparison between the observational data and the efhi thresholds predicted by a bi - maxwellian approach , which is also reproduced here in figure [ f5 ] by the dotted line corresponding to @xmath2 .
moreover , in figure [ f5 ] we show that this disagreement may be resolved by the instability thresholds derived for bi - kappa models which are more appropriate to describe the suprathermal electrons .
the instability thresholds are markedly changed with decreasing the power index @xmath5 and for lower values of @xmath5 these thresholds are approaching the limits of the temperature anisotropy observed in the solar wind .
what we found even more interesting is that the instability thresholds also shape very well the isocontours of the observational data , counting the number of events in the bottom panel .
developing for large temperature anisotropies exceeding these thresholds , the fhi dissipates the free energy and enhances the electromagnetic fluctuations , which may also prevent the anisotropy to grow by scattering particles back towards quasi - equilibrium states .
given that suprathermal electrons in the solar wind are practically collisionless , such a good agreement between the instability thresholds predicted by the kinetic theory and the limits of the temperature anisotropy reported by the observations represents an important confirmation on the role played by the fhi instability in the relaxation process .
.fitting parameters for thresholds @xmath69 [ cols="^,^,^,^,^,^,^,^",options="header " , ]
in this paper we have proposed a refined theory of the electron firehose instability in anisotropic kappa distributed plasmas , which provide a new and , in our opinion , valuable evidence of an extended implication of this instability in the relaxation of the temperature anisotropy in collision - poor plasmas from space .
our present study is particularly motivated by the solar wind observations which do not confirm the indefinite increase of temperature predicted by the solar wind expansion in the direction parallel to the interplanetary magnetic field , but reveal very clear bounds for the temperature anisotropy of plasma particles .
previous studies have focused to the thermal ( core ) populations of electrons and protons , using standard bi - maxwellian approaches , and have shown that large deviations from isotropy are constrained by the kinetic instabilities @xcite .
however , the same bi - maxwellian is not appropriate to describe suprathermal populations and their anisotropy , and can not prescribe accurately the resulting instabilities and their back reaction on these populations .
here we have assumed the anisotropic electrons well reproduced by the bi - kappa distribution function , which is the empirical model invoked by @xcite to describe the velocity distribution of suprathermal electrons in the solar wind .
in addition , the temperature of the suprathermal population is considered dependent on the power - index @xmath5 , enabling us a realistic interpretation of the suprathermals and their effects ( theoretical and observational arguments are detailed in the introduction ) .
the results of our present study contrast with those provided by @xcite and @xcite , who studied the same efhi but driven by bi - kappa electrons with a @xmath5-independent temperature .
these differences are made clear in the discussion from this section .
two distinct regimes of the efhi are identified in section 3 , and these regimes are differentiated by the wave - number dispersion laws ( curves ) obtained for the wave - frequency and growth - rate of the instability .
thus , more specific to the efhi are the unstable lh - polarized modes exemplified in figures [ f1 ] and [ f2 ] with frequencies that can significantly exceed the proton cyclotron frequency @xmath48 .
for this regime to be established , the kinetic free energy of the anisotropic electrons must be sufficiently large , and this usually means a high enough plasma beta or a large anisotropy .
if damped , these modes can not extend above @xmath48 and their wave - number dispersion keeps the aspect of low - frequency emic modes in the absence of kinetic anisotropies . at higher wave - numbers ( lower scales ) these damped modes can change their polarity converting to the branch of rh - polarized modes ( whistlers ) .
these electromagnetic modes with a wave - number dispersion resembling that of the emic modes , i.e. , with wave - frequency increasing asymtotically to @xmath48 , can be driven unstable by the efhi for conditions approaching marginal stability .
a few cases relevant for this new regime are presented in figures [ f3 ] and [ f4 ] , with mention that top panels include unstable solutions representative for a transition between these two distinct branches of the efhi .
we should observe that considering plasma parameters with values typical for the solar wind conditions , e.g. , in figures [ f1]-[f4 ] , the efhi develops only in the presence of suprathermal electrons , i.e. , for finite values of @xmath5 , while for ( bi-)maxwellian limit @xmath2 these modes are damped .
increasing the presence of suprathermal populations ( by lowering @xmath5 ) has opposite effects on the wave - frequency of the unstable modes , which become evident if we compare for instance figures [ f2 ] and [ f4 ] .
however , the efhi is clearly stimulated by the suprathermal electrons , which enhance the ( maximum ) growth - rates in both these two regimes .
noticeable is the significant increase shown by the growth - rates for conditions approaching the marginal stability ( figure [ f3 ] ) , which can also explain the significant decrease of the instability thresholds shown in figure [ f5 ] .
these thresholds are markedly lowered with decreasing @xmath5 , and for lower values of @xmath5 they shape very well the limits of temperature anisotropy reported by the observations in the solar wind . in the previous studies involving bi - kappa electrons with a @xmath5-independent temperature , e.g. , in @xcite , the existence of these two distinct regimes was not mentioned , and , in contrast to our present results , the suprathermals were found inhibiting the efhi , and departing the instability thresholds from the limits of temperature anisotropy in the solar wind .
our present results strongly suggest that the efhi may efficiently constrain the temperature anisotropy of the suprathermal electrons in the slow wind , complementing the results by @xcite , which showed the same effects of this instability on the core electrons .
a good agreement between the instability thresholds and the bounds of the temperature anisotropy measured in the solar wind is conditioned by a proper modelling of the velocity distributions in accord to the observations . in conclusion
, the efhi can be considered a plausible mechanism of electron energy transfer between the directions parallel and perpendicular to the uniform magnetic field . from an extended perspective
, we can further claim that the resulting low - frequency fluctuations can establish an energy transfer from small to large scales , namely from the electrons , especially the energetic or suprathermal electrons which carry the main heat flux in the solar wind , to the resonant protons . although suprathermal populations are not easily captured in numerical experiments , it becomes however clear that our present results provide valuable premises that may stimulate new and advanced simulations to confirm these mechanisms .
the authors acknowledge support from the katholieke universiteit leuven , ruhr - university bochum , and alexander von humboldt foundation . these results were obtained in the framework of the projects goa/2015 - 014 ( ku leuven ) , g0a2316n ( fwo - vlaanderen ) , and c 90347 ( esa prodex 9 ) .
the research leading to these results has also received funding from the european commission s seventh framework programme fp7-people- 2010-irses-269299 project - solspanet ( www.solspanet.eu ) .
shaaban would like to thank the egyptian ministry of higher education for supporting his research activities .
pierrard v. , lazar m. , tverk , .
, poedts s. , maksimovic m. , trvnek p. , 2016
, the electron temperature anisotropy in the solar wind . i. a comparative analysis of the core and halo populations .
2016 , ( submitted ) | the indefinite increase of temperature predicted by the solar wind expansion in the direction parallel to the interplanetary magnetic field is already notorious for not being confirmed by the observations . in hot and
dilute plasmas from space particle - particle collisions are not efficient in constraining large deviations from isotropy , but the resulting firehose instability provides in this case plausible limitations for the temperature anisotropy of the thermal ( core ) populations of both the electron and proton species .
the present paper takes into discussion the suprathermal ( halo ) electrons , which are ubiquitous in the solar wind .
less dense but hotter than the core , suprathermals may be highly anisotropic and susceptible to the firehose instability .
the main features of the instability are here derived from a first - order theory for conditions specific to the suprathermal electrons in the solar wind and terrestrial magnetospheres . unveiled here
, new regimes of the electron firehose instability may be exclusively controlled by the suprathermals .
the instability is found to be systematically stimulated by the suprathermal electrons , with thresholds that approach the limits of the temperature anisotropy reported by the observations .
these results represent new and valuable evidences for the implication of the firehose instability in the relaxation of the temperature anisotropy in space plasmas .
[ firstpage ] sun : solar wind electrons temperature anisotropy electromagnetic instabilities methods : analytical ; observational |
the extreme - mass - ratio inspirals ( emris ) of compact objects ( white dwarfs , neutron stars or stellar - mass black holes ) with mass @xmath1 into massive black holes with mass @xmath2 can serve as excellent probes of strong - field general relativity ( see @xcite and references therein for details ) .
the detection of gravitational waves from emris is one of the main targets for the planned gravitational - wave detector lisa ( laser interferometer space antenna ) @xcite .
astrophysical uncertainties mean that rate predictions are somewhat uncertain , but tens to thousands of emris may be observed over the duration of the lisa mission @xcite .
detection of emris is not an easy task , however .
the expected gravitational wave ( gw ) signals from emris will be buried in instrumental noise and a foreground of galactic white - dwarf binaries .
the instantaneous signal amplitude will typically be an order of magnitude smaller than the amplitude of noise and foreground fluctuations in the detector at the same frequency .
matched filtering allows the identification of weak sources buried in noise , but this relies on the generation of templates of possible signals present in the data .
the emri parameter space is fourteen dimensional .
following the notation of barack and cutler @xcite , these parameters are : the central black hole mass , @xmath3 , the mass of the inspiraling object , @xmath4 , the dimensionless spin parameter of the central black hole , @xmath5 , the orbital frequency , @xmath6 , and orbital eccentricity , @xmath7 , specified at some moment of time , @xmath8 , the inclination of the orbit to the equatorial plane of the central black hole , @xmath9 , the ecliptic latitude , @xmath10 , and longitude , @xmath11 , of the source on the sky , the orientation of the spin axis of the central black hole , @xmath12 and @xmath13 , the distance to the source , @xmath14 , and three phase angles , @xmath15 , @xmath16 and @xmath17 , which specify , respectively , the azimuthal orbital phase , periapsis precession phase , and phase of the orbital - plane precession at @xmath18 .
the large dimensionality of the emri parameter space , combined with the long observation timescale ( @xmath19 year ) , makes fully coherent matched filtering using a template bank impractical as a detection method @xcite .
time - frequency searches are one possible alternative to coherent matched filtering .
these have a much lower computational cost , albeit at the price of lower sensitivity .
time - frequency methods consist of building a spectrogram of the signal by dividing it into shorter segments and performing a fourier transform on each of these segments , identifying possible tracks in the spectrogram , and then finally estimating source parameters from the tracks .
we previously used time - frequency searches to analyse the mock lisa data challenge ( mldc ) @xcite round 2 data set @xcite , and found that we were able to get good parameter recovery . that previous analysis used the hierarchical algorithm for clusters and ridges ( hacr ) @xcite to identify tracks by clustering bright pixels in binned spectrograms , followed by a simple parameter estimation routine that used the track shape to infer the six intrinsic emri parameters @xmath20 and used the power variation along the tracks to estimate the two sky position angles @xcite .
the hacr algorithm looks for clusters of any shape , and uses a binning technique that is based on rectangular pixel bins @xcite .
this is a rather inefficient way to search for emri tracks , which have a characteristic chirping shape .
the most recent release of the mldc , round 1b , was a repeat of the type of emri data previously released in round 2 @xcite .
for this analysis we developed and used an improved track detection algorithm , which is tuned to look for emri type signals .
we call this algorithm the chirp - based algorithm for track search ( cats ; see @xcite for additional details ) . for the round 1b analysis
, we also applied parameter - estimation methodology that had been improved relative to the techniques used in round 2 @xcite .
the improvements included an estimation of the inclination of the source to the line of sight and a metropolis - hastings monte carlo search to find the parameters that best reproduced the detected tracks .
the mldc round 1b data set consisted of five separate emri challenge data sets , 1b.3.1 - 1b.3.5 ; each challenge set comprised a single emri signal embedded in noise , with signal - to - noise ratio ( snr ) between 40 and 110 @xcite .
we were able to detect emri signals in all of these data sets , and estimated 9 parameters for each challenge data set : the six intrinsic parameters @xmath21 and @xmath6 , the two sky direction angles , @xmath10 and @xmath11 , and the angle @xmath22 between the smbh spin vector and the line of sight to the source .
we also independently estimated the plunge time , but this is determined by the other intrinsic parameters and was used to improve the determination of those parameters .
overall , the parameter recovery from the time - frequency techniques was very good and better than achieved in the round 2 analysis @xcite .
the paper is organised as follows . in section [
sec : search ] we describe the new track detection techniques and in section [ sec : estimate ] we describe the improvements to our parameter estimation algorithms . in section
[ sec : results ] we report the results of our analysis of the round 1b data and compare these to the true parameters of the challenge data sets .
section [ sec : future ] includes a summary and a discussion of possible future improvements to the search pipeline .
our track search for the mldc round 2 analysis used hacr @xcite .
we made a first cut at the round 1b analysis using the same technique .
we were able to detect several bright tracks for sources 1b.3.1 , 1b.3.2 and 1b.3.3 , and a cluster for 1b.3.4 .
the 1b.3.4 detection was very noisy , which was disappointing since this was a high snr source .
this failure prompted us to look at improved techniques more tuned to emri detection .
hacr does not use all of the available information about the signal . specifically
, we expect the tracks in the time - frequency spectrogram to be chirping curves characterized by three parameters : the frequency @xmath23 and its first two time derivatives @xmath24 and @xmath25 .
( the third time derivatives of frequency along the tracks are vanishingly small for low - eccentricity signals like those in challenges 1b.3.1 - 1b.3.3 , and can be neglected for challenges 1b.3.4 and 1b.3.5 if sufficiently short sections of the tracks are considered . ) moreover , the first and second time derivative of frequency must be positive .
this additional information can help to detect tracks .
in fact , when we scanned the time - frequency spectrogram by eye , we found some tracks that hacr did nt report , because we intuitively apply this additional information about the expected track shape when performing a visual search for tracks . to make use of this information
, we built a chirp - based algorithm for track search ( cats ; see @xcite for additional details ) to aid in track detection .
this algorithm works as follows : * construct @xmath26 and @xmath27 time delay interferometry ( tdi ) channels for the challenge data sets . * for each channel ,
construct a spectrogram by dividing the data into time segments of equal duration , then fourier transforming the data in each time segment after multiplying it by a hanning window to reduce edge effects .
the time segment durations were @xmath28 seconds for challenge 1b.3.1 , @xmath29 seconds for challenges 1b.3.2 and 1b.3.3 , and @xmath30 seconds for challenges 1b.3.4 and 1b.3.5 .
* whiten the two spectrograms by dividing the signal power in each pixel by the expected lisa noise power spectral density , then construct a joint spectrogram by adding the two normalized spectrograms . *
define the starting and ending time of a chirp search .
construct a three - dimensional grid of parameters in the @xmath31 space .
for each point in the parameter space , build a potential track . then add up the power in all pixels along this track between the starting and ending time of the search to obtain the total track power .
( this is , effectively , a variation on a template - based matched - filtering search in the @xmath31 space . ) * find the brightest track and claim it as a detection . set the power in all pixels along this track in the spectrogram to large negative values in order to keep
future detected tracks from intersecting the current track .
repeat this step until all tracks are detected . for this round ,
we did not attempt to search the whole spectrogram using this technique , nor did we search over the entire parameter space of @xmath31 .
instead we used the hacr algorithm to identify clusters in the spectrogram arising from the presence of signals in the data stream , and the approximate shapes of these tracks .
we then targeted those areas for follow - up with the cats algorithm . at the time of the round 1b analysis
, we had not robustly determined the correct thresholds for track detection . in the future
, the track power threshold will be set based on a desired value of the false alarm probability . for this round ,
we set thresholds by hand , but performed a number of `` sanity checks '' to decide whether a given track detection was to be believed .
these checks included : * compare the detected tracks with a visual observation of the spectrogram .
* check if the tracks have sidetracks due to fft bleeding , i.e. , tracks that are parallel to them but one pixel off on each side . *
check if there is a clear harmonic structure typical to emris , with sidebands caused by orbital - plane precession , i.e. , several tracks that are nearly parallel ( same @xmath24 ) with a fixed offset from each other .
* check if the track parameters are reasonable : e.g. , if presumed sections of a track are found at early and late times , the frequency and frequency derivative at the later time should be greater than those at the earlier time .
in all the round 1b challenge data sets , cats successfully found all tracks that were detected by hacr or could be made out by eye .
for some of the challenges , cats found additional sidebands and cats was also able to find several tracks in the round 1b.3.5 data set , in which hacr was unable to identify any interesting clusters .
this was particularly pleasing , since previous work @xcite had indicated that sources of low snr , like 1b.3.5 , would be out of reach of time - frequency methods .
we estimated the systematic uncertainties of cats by comparing the results of track searches using different time steps and different gridding of the chirp parameter space .
these comparisons suggest uncertainties on the order of @xmath32 hz , @xmath33 hz / s , and @xmath34 hz / s@xmath35 . of course
, the inherent errors in chirp parameter estimation of low - snr signals , particularly for short tracks , could be much greater than these systematic uncertainties .
we have also been developing and testing another time - frequency technique that utilizes the modified radon transform . for this search
, we sum power in pixels along expected tracks in the time - frequency spectrogram of a given source .
the expected tracks are calculated by integrating the known evolution equations for emris for a given set of initial parameters .
tracks of the harmonics and sidebands of orbital frequencies are included in the summation , and we search through the whole parameter space for the maximum signal - to - noise ratio .
this technique was not fully developed at the time of submission , so although the results were compared to the results obtained from the cats algorithm , only the latter results were submitted for this round . in the future ,
further improvement of the radon transform algorithm will be to needed in order to overcome problems of getting stuck in local maxima of the parameter space .
we are also investigating similar techniques using the hough transform which sums over number counts ( instead of powers ) of bright pixels along the track . a search using the radon or
hough transform normally takes minutes to complete on a laptop when a good initial guess has been made , or hours starting from a poor guess .
parameter extraction based on the tracks identified by cats consisted of two steps . in the first step we made rough estimates of the parameters .
this was the approach used for parameter extraction in our analysis of the round 2 data sets @xcite .
an emri waveform is characterized by three fundamental frequencies the orbital frequency , @xmath36 , the perihelion precession frequency , @xmath37 and the frequency of precession of the orbital plane , @xmath38 .
each track identified in the spectrogram is a harmonic of these three frequencies characterized by three integers , @xmath39 , with @xmath40 .
the analytic kludge ( ak ) waveforms of barack and cutler @xcite , which is the model used to generate the mldc data sets , are quadrupole in nature and precession effects are included in an ad hoc way by precessing the observer about the source .
this restricts @xmath41 and , in practice , harmonics with @xmath42 are suppressed relative to the @xmath43 harmonics and are not detected by a time - frequency analysis . if three tracks are detected at a given time
two @xmath44-harmonics plus a @xmath45 sideband for one of those harmonics , then the separation of the @xmath44 harmonics is a measure of @xmath36 , the separation of the @xmath45 sideband is a measure of @xmath46 and , if the harmonic numbers of one of the harmonics can be guessed , the frequency of that harmonic then gives @xmath47 . for a specification of @xmath7 and @xmath9
, @xmath48 determines @xmath5 , @xmath49 determines @xmath3 and @xmath6 is a parameter that has been measured directly .
the rate of change of frequency of one harmonic , @xmath24 , then determines @xmath4 . if the plunge time and the frequency of at least one harmonic at some other time can also be estimated , it is possible to iterate on @xmath7 and @xmath9 to determine all six intrinsic parameters for the source .
the motion of the lisa detector around the sun leads to a modulation in the detector response that depends on sky position . from the power variation along a track
it is therefore possible to estimate the sky position of the source , @xmath50 by using the low - frequency approximation to the lisa response @xcite . for the round 1b analysis
, we also tried to use the relative power in the sidebands to estimate the orientation of the source relative to our line of sight .
physically , the relative sideband power can only depend on the inclination of the source to the line of sight , @xmath22 , which is given in terms of the standard waveform parameters by @xmath51 .
the relative harmonic power depends only on the absolute value of @xmath52 , and so we can choose @xmath53 $ ] .
it is not possible to determine the two source orientation angles , @xmath54 , from the single number @xmath22 and so for the mldc entry we returned @xmath22 as an additional parameter .
it turns out that , for a fixed value of @xmath44 , the relative power in different @xmath45 harmonics depends only on @xmath22 and the inclination of the orbit , @xmath9 ( more details will be given in @xcite ) .
the power ratio in sidebands can thus be used to estimate @xmath22 and also to improve the determination of @xmath9 .
this is illustrated in figure [ kapest ] .
the curves in this figure show the power in the various sidebands relative to the power in the @xmath55 harmonic as a function of @xmath22 , for a fixed value of @xmath56 .
the horizontal dashed line shows the power ratio @xmath57 estimated from the spectrogram , from which we infer an estimate @xmath58 . in the analysis of the round 1b data
, we found @xmath59 for all the data sets ( specifically @xmath60 for challenges 1b.3.11b.3.5 respectively ) .
this suggested the determination of @xmath22 was weak .
the true values were @xmath61 respectively , which implies the measurement was quite accurate , except for 1b.3.1 in which the sideband detected was very weak , and so the harmonic power ratio was noise dominated .
the value @xmath62 seems to be typical for points drawn from the mldc priors , and the t - f spectrogram has limited ability to pin @xmath22 down more accurately .
however , the power ratio did help to improve the estimation of @xmath9 for these data sets . after determining rough parameter estimates via the above methods , the second stage of parameter estimation , which was used for the first time in this round of the mldc , was to refine the estimates using an implementation of the metropolis - hastings monte carlo ( mhmc ) algorithm .
the mhmc algorithm is described in many standard textbooks ( e.g. , @xcite ) and is being used extensively for the detection of various sources in the mldc @xcite .
an mhmc search constructs a sequence of points in parameter space , @xmath63 . from a point @xmath63 ,
a value for the next point in the chain is proposed by drawing from a proposal distribution @xmath64 .
a random parameter is then drawn from a uniform distribution @xmath65 $ ] and if @xmath66 the move is accepted and the chain moves to @xmath67 , otherwise the move is rejected and @xmath68 .
here @xmath69 is the metropolis - hastings ratio @xmath70 @xmath71 is the prior on the parameters and @xmath72 is the likelihood of the data @xmath73 given the parameters @xmath74 . for the time - frequency mhmc search
, we used priors that were uniform within the parameter ranges specified for the mldc , and based the log - likelihood on the overlap between the observed whitened time - frequency spectrogram and a theoretical spectrogram built from the ak model ( see @xcite for additional details ) .
our proposal distribution was constructed from the fisher information matrix for our waveform model . for the round 1b analysis ,
the mhmc chains were seeded at the rough parameters estimated in the first stage and were allowed to run for @xmath75 points .
the search used plunge eccentricity and plunge time as parameters rather than initial frequency and eccentricity , with a prior on the plunge time estimated from the detected time - frequency tracks .
the mhmc chains refined the initial parameter estimates , but did not move far from the initial parameters , as is clear by comparing the `` rough '' and `` final '' parameters in table [ table : tfparams ] .
the chains moved somewhat further for 1b.3.5 , for which a mis - identification of the harmonics led to a poor initial parameter estimation .
we also ran chains that started at random points in the parameter space , and these converged to the same parameter estimates , which gave us confidence that the code was working as expected .
in table [ table : tfparams ] we tabulate the true and recovered parameters for our search of the mldc round 1b data . in the table
`` rough '' denotes the first , rough parameter estimates , `` final '' denotes the refined estimates obtained from the mhmc search and `` true '' denotes the true parameters .
we see that , overall , we obtained good estimates of all the parameters , except for the lowest snr challenge , 1b.3.5 .
the `` rough '' parameter estimation was already quite good in most cases , but the parameter estimation was further improved by the mhmc , except in the case of challenge 1b.3.1 where the final parameters were worse than the rough parameters .
our sky position estimator has a degeneracy between antipodeal sky positions @xmath50 and @xmath76 , which reflects the real degeneracy between these points in the low frequency approximation @xcite .
we used the convention that we returned values with @xmath77 $ ] , which meant that we returned the antipodeal position for sources 1b.3.4 and 1b.3.5 .
the sky position estimates were otherwise very good .
we make some source specific comments below . .time - frequency search results for challenges 1b.3.1 - 1b.3.5 .
[ cols="^,^,^,^,^,^,^,^,^,^",options="header " , ] the high smbh mass meant that there were a number of near - degeneracies in the parameter space for this challenge , which made individual parameters difficult to resolve despite the high snr .
this problem was exacerbated by the fact that the inclination was such that only one bright @xmath45-sideband was seen for each @xmath44-harmonic .
although other , weak , sidebands were detected , the mhmc moved towards lower eccentricity and inclination since this put more of the signal power into the brightest harmonic and hence `` improved '' the fit . in a future implementation of the search , we plan to include a penalty associated with parameter space points that do not reproduce all the tracks that have been seen .
the final parameters of the mhmc search were such that the strongest detected track for the @xmath78 harmonic should be the @xmath79 sideband , and the secondary track should be the @xmath80 sideband . however , the values of @xmath9 and @xmath22 were such that the @xmath81 sideband should have been brighter than the @xmath80 sideband
. since we could not detect another sideband in the presumed location of @xmath81 , we manually changed the inclination angle @xmath22 , which decouples from the other parameters , to get a fit to the detected tracks .
the resulting estimates for @xmath9 and @xmath22 were quite poor as a result .
this was the only challenge in which the rough parameter estimates were better than the final parameter estimates .
in future searches , we may use only the rough parameter estimation for high mass smbh sources as it is then easier to control degeneracies .
the mhmc search pushed the smbh mass all the way to the maximum value allowed by the prior , which suggested a lack of accuracy .
the initial rough parameter estimate matched the data well , although not quite as well as the final mhmc result .
we submitted both results , although this was artificial in a sense , because we felt the final point was a better fit and only thought it was wrong because we had artificially tight prior constraints on @xmath3 .
when the prior was relaxed , the best fit was found to be for a slightly higher central black hole mass , around @xmath82 .
again , the mhmc search pushed the smbh mass to the maximum value allowed by the prior .
this was , however , very close to the value detected by the initial rough estimate . once again , if a less restricted prior was used , the best fit would have been for a slightly higher central black hole mass .
for both 1b.3.2 and 1b.3.3 , the true black hole mass was quite close to the edge of the prior which is why this issue arose .
although sections of multiple harmonics and sidebands were detected , the duration of each detected segment was quite short .
this suggested that parameter estimation might not be as accurate as usual .
however , the final parameters reproduced the detected tracks well , which gave us confidence in the results , and it is clear from table [ table : tfparams ] that parameter recovery was very good .
the tracks were difficult to find because of the low snr . nevertheless , we succeeded in finding a track corresponding to the @xmath83 harmonic with a possible sideband near @xmath84 , and tracks corresponding to the @xmath78 and @xmath83 harmonics , each with one sideband , near @xmath85 s. the short duration of the tracks made accurate parameter estimation difficult , and the best fit parameters did not yield tracks that passed through the tracks detected near @xmath84 ( although they were close ) .
we expected these parameters to be the worst of the five challenges , which is confirmed by table [ table : tfparams ] .
however , we were unable to detect anything in the corresponding data set in mldc round 2 , so the performance of our approach in round 1b is a significant improvement .
the time - frequency analysis of the round 1b data was very successful and our entry had the most complete parameter recovery out of the three submissions for this challenge .
the other challenge entries used mhmc matched filtering
. ultimately that will be more sensitive and have much better parameter recovery than time - frequency analysis , but the time - frequency approach clearly has some parameter estimation power and will potentially be a useful first step in an analysis pipeline .
there are a few ways in which the search could be improved . the cats algorithm , as applied to round 1b , was not automatic , but interesting regions of parameter space were selected either by eye or by using the output of a hacr search . in the future
, we will assess the performance of the cats algorithm without using prior information .
we will also compare the radon transform algorithm outlined earlier against the cats algorithm .
the parameter extraction can not be significantly improved .
however , the template spectrogram model used in round 1b to evaluate the likelihood in the mhmc search is only an approximation .
it needs to be properly compared against mock data and might be improved .
we also want to investigate imposing penalties in the likelihood evaluation to ensure the parameters predict all the tracks that are detected in the spectrogram .
it should also be possible to use the total power in the tracks to estimate the total snr or equivalently the distance to a source , which we have not done so far .
the parameter estimation that we have done to date has been very specific to the ak model used for the mldc data .
the general techniques will generalise to more accurate waveform models but the mapping between harmonic frequencies and physical parameters will change .
we plan to derive analogous methods for parameter extraction from `` numerical kludge '' waveforms @xcite as a first step towards this non - trivial generalisation .
so far , the mldc emri data sets have each contained a single emri signal in instrumental noise .
the prior that there is only one signal in the data set is a very powerful one , but also very unrealistic .
the true lisa data stream will contain a few tens to as many as a thousand emri sources @xcite plus signals from white dwarf binaries and supermassive black hole mergers .
time - frequency detection algorithms , plus the simple parameter extraction techniques that we have used to date , will have limited effectiveness in such situations .
however , it is a useful exercise to see at what point these techniques no longer work , and it is likely that the methods will still be able to detect the brightest sources even in the presence of confusion , although track detection thresholds will have to be modified to account for the confusion `` noise '' . the mldc round
3 will begin to address these issues , as there will be only one emri data set containing five overlapping sources .
we will explore confusion while attempting to analyse round 3 .
however , the sources will also be low snr , so it is not yet clear whether time - frequency techniques will be able to detect anything . if necessary
, we will create similar data containing brighter sources .
we also plan to explore the efficiency of these methods when applied to data containing a single emri with a reduced galaxy to mimic the effect of confusion from white - dwarf binaries .
jg thanks the royal society for support and the albert einstein institute for hospitality and support while part of this work was being completed .
i m was partially supported by nasa atp grant nnx07ah22 g to northwestern university .
lw s work is supported by the alexander von humboldt foundation s sofja kovalevskaja programme funded by the german federal ministry of education and research .
10 amaro - seoane p , gair j r , freitag m , miller m c , mandel i , cutler c j and babak s , _ class .
quantum grav .
_ * 24 * r113 ( 2007 ) .
danzman k _ et al .
_ _ lisa pre - phase a report _
, report mpq 233 ( 1998 ) .
barack l and cutler c , _ phys .
d _ * 69 * 082005 ( 2004 ) .
gair j r , barack l , creighton t , cutler c , larson s l , phinney e s & vallisneri m , class .
* 21 * , s1595 ( 2004 ) .
arnaud k a _ et al .
_ , _ aip conf
_ , * 873 * 619 ( 2006 ) .
gair j r , mandel i and wen l. accepted to a special issue of the journal of physics conference proceedings , arxiv:0710.5250 ( 2007 ) .
gair j r and jones g , 1145 ( 2006 ) .
wen l & gair j r , * 22 * s445 ( 2005 ) .
r & wen l , * 22 * s1359 ( 2005 ) .
arnaud k a _ et al .
_ , s551 ( 2007 ) .
gair j r & mandel i , in preparation ( 2008 ) .
c , 7089 ( 1998 ) .
wen l , chen y & gair j r , _ aip conf .
_ , 873 , 595 ( 2006 ) .
gregory p , _ bayesian logical data analysis for the physical sciences _
( cambridge : university press ) ( 2005 ) .
gair j r & glampedakis k , 064037 ( 2006 ) .
babak s , fang h , gair j r , glampedakis k and hughes s a , 024005 ( 2007 ) . | the planned laser interferometer space antenna ( lisa ) is expected to detect gravitational wave signals from @xmath0 extreme - mass - ratio inspirals ( emris ) of stellar - mass compact objects into massive black holes .
the long duration and large parameter space of emri signals makes data analysis for these signals a challenging problem .
one approach to emri data analysis is to use time - frequency methods .
this consists of two steps : ( i ) searching for tracks from emri sources in a time - frequency spectrogram , and ( ii ) extracting parameter estimates from the tracks . in this paper
we discuss the results of applying these techniques to the latest round of the mock lisa data challenge , round 1b .
this analysis included three new techniques not used in previous analyses : ( i ) a new chirp - based algorithm for track search for track detection ; ( ii ) estimation of the inclination of the source to the line of sight ; ( iii ) a metropolis - hastings monte carlo over the parameter space in order to find the best fit to the tracks . |
recently , extended red emission ( ere ) has been detected in the diffuse galactic background and in high - galactic latitude cirrus clouds with surprisingly high intensities ( gordon , witt , & friedmann 1998 ; @xcite ) .
this emission , appearing in the 500800 nm spectral range as an unstructured broad feature , is attributed to photoluminescence by some component of interstellar grains .
the correlation between the ere r - band intensity and -column density of ( [email protected])10@xmath1 ergs s@xmath2 @xmath2 sr@xmath2 h atom@xmath2 ( @xcite ) implies that , at a minimum , ( 10@xmath03)% of all uv / visible photons in the diffuse interstellar radiation field are absorbed by the ere producing grains , if each absorption results in the emission of an ere photon .
this requires that the ere - grains make a significant contribution to the interstellar extinction in the uv / visible range , which further implies that they must consist of cosmically abundant elemental constituents in a form that permits extremely high efficiency photoluminescence . after a review of possible ere - grain candidates , witt , gordon , & furton ( 1998 ) and ledoux et al .
( 1998 ) independently proposed that silicon nanoparticles ( snps ) might be the source of ere in the interstellar medium in as far as they produce photoluminescence spectra closely matching those observed in ere sources and match the constraints regarding the required quantum efficiency and elemental abundance .
it was discovered experimentally that the photoluminescence emission originates from nanocrystalline si regions in porous si material ( @xcite ; @xcite ) and from clusters of nanosized grains ( @xcite ; @xcite ; @xcite ) .
the estimate of the amount of silicon needed in the form of snps , however , was based upon the average uv absorption cross section of crystalline silicon in bulk form .
this must be considered as only a rough approximation , because the remarkable optical properties of snps , clusters involving a few 10@xmath3 silicon atoms each with a silicon suboxide ( sio@xmath4 ) surface covering , arise precisely because quantum confinement alters the band structure of the cluster material relative to that in the bulk material .
since quantum confinement is a function of cluster size , so are the resulting photoluminescence characteristics , and so are the dielectric functions of the cluster material .
the current investigation was undertaken to answer several related questions : what are the uv / optical extinction characteristics of a separate component of snps ?
what fraction of the silicon believed to be depleted into solid grains is needed in the form of snps to explain the observed ere intensities ? what resulting characteristics of the interstellar extinction curve can be related to absorption by snps and can thus be correlated with varying ere intensities ? to find answers on these questions , we made use of a powerful computational approach based on the method of regularization ( @xcite ) , which was proven to be efficient in modeling of interstellar dust ( zubko , kreowski , & wegner 1996 , 1998 ) .
the first step in our study was to choose the optical constants of snps .
witt et al . ( 1998 ) proposed that the interstellar snps should have diameters of 1.55 nm because only grains in this size range exhibit efficient photoluminescence .
evidently , the dielectric properties of such snps should be different from those of a bulk material because of both the quantum size confinement effect and the increased role of surface states .
these expectations are fully confirmed by available experimental data ( @xcite ; @xcite ) . in particular , thei ( 1997 ) reported the dielectric function of nanosized silicon from porous samples with a range of porosities 6080% , corresponding to the respective range of crystallite sizes of 52 nm .
these results show that the optical properties of silicon nanostructures have a strong size dependence .
however , an accurate derivation of the latter is not a trivial problem , because in each silicon sample there is some size distribution of crystallites ( @xcite ) .
note that thei ( 1997 ) dielectric function covers the wavelengths 0.21 @xmath5 m only .
however , as we will see below , we especially need the data down to the lyman limit ( 912 ) .
we therefore decided to use in our modeling the optical constants of nanosized silicon by koshida et al .
( 1993 ) spanning the range 0.0550.85 @xmath5 m , which is sufficient for our problem .
these constants correspond to the characteristic diameter of a snp of around 3 nm .
the optical constants of nanosized silicon by koshida et al .
( 1993 ) and of bulk silicon taken from the compilation by palik ( 1991 ) are displayed in fig .
[ fig : oc ] .
there is a considerable difference between the two sets which may also be seen in the extinction efficiencies of the respective spherical grains shown in fig .
[ fig : qext ] .
the first three cases correspond to the snps used in our modeling ( see section 3 for more details ) .
curves 3 and 4 demonstrate the so - called _ extrinsic size effect _ for small particles when the bulk optical constants are used .
the peak at 0.118 @xmath5 m is due to the sio@xmath6 shell and is present also in the spectra of bulk material , whereas the peaks at 0.153 @xmath5 m and 0.138 @xmath5 m ( for 3 and 4 , respectively ) are surface plasmons due to the silicon core , specific for small particles .
the peak position depends on the volume fraction of si .
in contrast to the latter cases , curve 1 shows the _ intrinsic size effect _ when the optical constants themselves vary with grain size .
note that draine & lee ( 1984 ) already computed the size - dependent dielectric function of small graphite grains .
however , the size dependence was taken into account in their calculation only for the free - carrier part of the dielectric function , whereas the part due to the interband transitions was set to be the same as for bulk graphite .
in contrast to draine & lee ( 1984 ) , the optical constants of nanosized silicon used in the present study explicitly reflect the drastic changes in the band structure and interband transitions of the snps in comparison with those of bulk silicon .
the dust models investigated for the present study were expected to satisfy three independent constraints simultaneously : 1 ) fit the mean galactic extinction curve as closely as possible over the entire near - ir to far - uv range ; 2 ) consume depleted heavy elements consistent with the recently proposed reference abundances for the interstellar medium ( about 2/3 of solar values , see e.g. , snow & witt 1996 ) ; and 3 ) require the ere - producing snps to absorb at least [email protected] of the uv / visible photons in the interstellar radiation field , as suggested by the observations of gordon et al .
the problem was solved by using the approach and computer program based on the method of regularization ( @xcite ; @xcite ) which allows one to find the size distributions of the dust components in the most general form .
the size distributions may be subject to the cosmic abundance and mass fraction constraints .
following zubko et al .
( 1998 ) , we treated the mixtures of spherical bare , multilayer and/or composite grains , whose extinction efficiencies were calculated either with the standard mie theory for bare and composite grains ( bohren & huffman 1983 ) or with its extension to multilayer particles . for composite grains the effective dielectric function was first calculated using the effective medium approach by stognienko , henning , & ossenkopf ( 1995 ) .
the fraction of uv / visible photons absorbed by snps , @xmath7 , was calculated with the formula @xmath8 where @xmath9 ( @xmath10 ) is the albedo of the entire dust mixture ( snps ) , @xmath11 ( @xmath12 ) is the extinction of the entire dust mixture ( snps ) , @xmath13 is the intensity of the interstellar radiation field at the earth in units of photons @xmath14 s@xmath2 @xmath2 .
the @xmath9 was taken from gordon et al .
( 1998 ) , @xmath11 , the mean galactic extinction ( @xmath15=3.1 ) from cardelli , clayton , & mathis ( 1989 ) and @xmath13 from mathis , mezger , & panagia ( 1983 ) .
the values of @xmath10 and @xmath12 were taken from the models .
we based our calculations on the g , m and o types of models introduced by zubko et al .
a g model contains the constituents usual for greenberg s models : the bare graphite grains and the silicate core organic refractory grains ( e.g. li & greenberg 1997 ) .
an m model is a mixture of the bare graphite and silicate grains together with the porous composite grains , resembling recent mathis ( 1996 ) models .
the constituents of an o model are the bare graphite and silicate grains together with the multilayer grains consisting of a silicate core coated by refractory organic and water - ice shells .
initially , we tried to model the extinction using the above models to which we added snps of 3.0 nm diameter covered by a sio@xmath6 mantle .
we varied the thickness of the mantle and found that the best results were achieved when the mantle is quite thin .
however , all the models resulted in quite low @xmath7 , as a rule not exceeding 0.030.04 , and , as a rule , consumed the solar fractional amount of silicon , 35 ppm ( atoms per 10@xmath16 h atoms ) .
for the next step we introduced spherical porous clusters of snps instead of isolated snps .
this resulted in an increase of @xmath7 , typically by 0.0030.005 ( the highest values correspond to 70% porosity ) .
then we tried combining the g , m , and o models and found that the only mixture that results in acceptable @xmath7 ( @xmath17 0.07 ) is a combination of g and o models or the go model .
it consists of four components : bare graphite grains , silicate core - organic refractory mantle grains , three - layer silicate - water ice - organic refractory grains and , finally , the snps in the form of either silicon core - sio@xmath6 mantle grains or 70%-porous aggregates of silicon grains containing a small amount of sio@xmath6 .
the go model requires a mass fraction of 7% of silicon grains .
it results in @xmath7=0.071 and 0.081 for the models with isolated and clustered snps , respectively .
note , however , that these models still consume much carbon , 190 ppm , and a solar fractional amount of silicon .
the go model seems more natural than g and o models separately , because when the dust grains in the interstellar medium do evolve through the formation and processing of mantles , we should expect the contributions into extinction of various intermediate stages .
the main results of our calculations are presented in fig .
[ fig : models ] and tables [ tab : tab1][tab : tab2 ] .
first of all , we see that the models with bulk silicon exhibit rather strong features in the uv ( see fig .
[ fig : qext ] ) which so far have not been observed . for this case
we adopted the volume fractions : @xmath18=0.297 and @xmath19=0.703 corresponding to the initial assumption of witt el al .
( 1998 ) about an equal amount of silicon and oxygen atoms in a snps .
by contrast , the contribution of the snps to the extinction is featureless when the optical constants of nanosized silicon and a very thin sio@xmath6 mantle are used . as a result ,
such models fit the mean galactic extinction much better .
practically all of the models require much carbon : 180190 ppm , except o models for which c / h=100 ppm , which is about optimal . as found in previous models by zubko et al .
( 1998 ) , more oxygen is needed for the o models .
the amount of silicon needed requires solar relative abundances .
a remarkable feature of the go models in contrast to the other models is the extremely small amount of magnesium and iron consumed , about 17 ppm , which is quite close to the respective optimum values by snow & witt ( 1996 ) .
in addition to providing the highest @xmath7 values , the go models also show the closest fit to the observed mean galactic extinction curve .
a result found to be generally true for all of our models is that @xmath7 is approximately equal to the mass fraction incorporated in the snps ( see table [ tab : tab2 ] ) .
this result is largely independent of whether the snps are included as a single - size component of @xmath20=1.5 nm , an @xmath21 distribution of porous snp clusters in the range 1.550 nm , or , as pointed out by the referee , if snps are included as the small - size tail of an @xmath22 distribution , as long as their mass remains constant .
since the cosmic abundance of si limits this mass fraction severely and approximately 50% of the available si is already incorporated in the snps in the go models , one important conclusion to draw is that if ere is produced by snps , the intrinsic photon conversion efficiency must be extremely high , essentially near 100% .
one reason why the @xmath7 values from table [ tab : tab2 ] are as small as they are is that the number of photons in the assumed radiation field is highest in the 0.350.55 @xmath5 m wavelength range , where the absorption contribution of the snps is practically zero .
one would obtain larger values of @xmath7 , if the interstellar radiation field contained increased numbers of uv photons .
if one can make a case for significant differences in the spectral energy distribution of the interstellar radiation field in different regions of the galaxy , one should be able to predict corresponding variations in the relative brightness of the ere and the visible diffuse galactic light ( dgl ) .
such variations should be apparent in results from ongoing surveys which map the dgl and the ere over a wide range of galactic longitudes ( e.g. , @xcite ) .
we conclude that in general the results of our models are consistent with the hypothesis that snps are the source of ere in the interstellar medium .
their extinction contribution occurs primarily at wavelengths shorter than 300 nm and is not expected to introduce discernable structure into the interstellar extinction curve , provided one takes into account the size dependence of the optical constants for nanostructured silicon .
the intrinsic photon conversion efficiency of such snps must be near 100% , if cosmic abundance constraints are to be met .
spherical clusters of snps with cluster radii up to 50 nm provide a slightly better match to the ere requirement than do single snps with the same total mass fraction .
it would be of great interest to predict possible spectral features due to snps in the infrared , as well as the contribution of snps in the total thermal emission of interstellar dust .
however , we can not do this because of the lack of optical constants of snps in the infrared and because of lacking information about the size distribution of snps . & & graphite & 22.1 & 143 & 0 & 0 & 0 & 0 & & sil - orr & 40.0 & 33 & 12 & 83 & 12 & 12 & & sil - ice - orr & 30.9 & 14 & 5 & 97 & 5 & 5 & & si - sio@xmath6 & 7.0 & 0 & 18 & 0 & 0 & 0
n & c & & 100.0 & 190 & 35 & 180 & 17 & 17 & & graphite & 22.1 & 143 & 0 & 0 & 0 & 0 & & sil - orr & 40.0 & 33 & 12 & 83 & 12 & 12 & & sil - ice - orr & 30.9 & 14 & 5 & 96 & 5 & 5 & & si - sio@xmath6-voids & 7.0 & 0 & 18 & 1 & 0 & 0 b & i & & 100.0 & 190 & 29 & 180 & 17 & 17 & & graphite & 23.5 & 145 & 0 & 0 & 0 & 0 & & sil - orr & 40.0 & 32 & 12 & 80 & 12 & 12 & & sil - ice - orr & 29.5 & 13 & 5 & 88 & 5 & 5 & & si - sio@xmath6 & 7.0 & 0 & 12 & 12 & 0 & 0 | to examine a recently proposed hypothesis that silicon nanoparticles are the source of extended red emission ( ere ) in the interstellar medium , we performed a detailed modeling of the mean galactic extinction in the presence of silicon nanoparticles .
for this goal we used the appropriate optical constants of nanosized si , essentially different from those of bulk si due to quantum confinement .
it was found that a dust mixture of silicon nanoparticles , bare graphite grains , silicate core - organic refractory mantle grains and three - layer silicate - water ice - organic refractory grains works well in explaining the extinction and , in addition , results in the acceptable fractions of uv / visible photons absorbed by silicon nanoparticles : 0.0710.081 . since these fractions barely agree with the fraction of uv / visible photons needed to excite the observed ere , we conclude that the intrinsic photon conversion efficiency of the photoluminescence by silicon nanoparticles must be near 100% , if they are the source of the ere . |
the properties of galaxies can vary depending on whether they reside in dense galaxy clusters or the field .
the most remarkable example of this is the morphology density relation ( dressler 1980 ) in which the proportion of elliptical galaxies increases toward the cores of rich clusters .
the morphology of galaxies in clusters has been based , mostly , on a visual classification scheme .
however , visual classification is only the first step in the characterization and description of galaxies .
it is necessary to conduct a quantitative morphological analysis of galaxies in clusters to answer basic questions like : are the properties of spiral galaxy discs , such as their scale - lengths , affected by the enviroment ?
such a study is also required to make a detailed comparison with , and therefore test , current theoretical predictions ( e.g. moore et al .
1999 ; gnedin 2003 ) .
the proximity and richness of the coma cluster has made it one of the most studied galaxy clusters . since godwin ,
metcalfe , & peach ( 1983 ) published the first wide - field galaxy catalog using photographic photometry , many other surveys have been conducted both in the central parts of this cluster ( e.g. , jorgensen & franx 1994 ; karachentsev et al .
1995 ; bernstein et al . 1996
; lobo et al.1997 ; secker & harris 1997 ; trentham 1998 ) and covering larger areas ( e.g. , kashikawa et al . 1995 ;
terlevich , caldwell & bower 2001 ; beijersbergen et al .
a recent survey combining wide field photometry and spectroscopy has been presented in komiyama et al .
( 2002 ) and mobasher et al .
there are numerous morphological studies of galaxies within the coma cluster , both in the optical ( e.g. , rood & baum 1967 ; dressler 1980 ; lucey et al . 1991 ;
jorgensen & franx 1994 ; andreon et al .
1996 ; andreon , davoust , & poulain 1997 ; gerbal et al .
1997 ; kashikawa et al.1998 ; mehlert et al .
2000 ; and komiyama et al . 2002 ) and in the near - infrared ( e.g. , pahre 1999 ; mobasher et al . 1999 ; khosroshahi et al .
2000 ) .
in this paper we present the morphology and structural parameters of galaxies in the central region of the coma cluster ( 0.28 square degrees ) .
we wish to stress that our analysis uses for the first time velocity data to establish cluster membership . furthermore , and importantly
, we do not a priori assume to know what the distribution of light is in elliptical galaxies or the bulges of spiral galaxies . that is , rather than force the @xmath9 model on these systems , we use srsic s ( 1968 ) @xmath0 model in an effort to _ measure _ the distribution / concentration of light . a detailed analysis of the relation between galaxy light concentration and galaxy environment was addressed in a previous paper ( trujillo et al .
2002a , hereafter t02a ) .
one of our present objectives is to study the various correlations among the structural parameters , and to search for possible differences according to morphological type or local conditions within the cluster .
a study of a rich and nearby cluster like coma is also very useful for establishing a local reference for studies of clusters at intermediate and high redshifts .
section 2 describes the observations and the compilation of redshifts .
the method to determine the quantitative morphology of galaxies is outlined in section 3 . in section 4
we explore the relationships between the structural parameters and also with the environment .
section 5 summarizes the main results of the paper .
we have performed a quantitative morphological analysis of galaxies in the coma cluster using a deep image taken in april 2000 with the wide field camera ( wfc ) at the 2.5 m isaac newton telescope ( int ) at the roque de los muchachos observatory .
the observations are described in detail in marn - franch & aparicio ( 2002 ) . here
, we outline the more relevant facts .
the image was taken through a sloan @xmath10-band filter , with a total exposure time of 3900 s ( 13@xmath11300 s ) .
the observations covered the inner 0.28 square degrees of the central part of the cluster ( see fig .
[ wfc ] ) .
conditions were photometric .
we performed a standard data reduction , comprising subtraction of bias , flat field corrections and co - addition of individual exposures .
each chip was calibrated using standard landolt stars ( landolt 1992 ) .
the seeing in the final ( combined ) image , measured using bright , unsaturated stars , was 1.1 arcsec .
the limiting magnitude for these observations was @xmath12 23.5 mag .
the four frames in fig .
[ wfc ] correspond to the four ( 2k @xmath11 4k ) ccds ( with a scale of 0.333 arcsecconds pixel@xmath7 ) of the wfc .
the position of the ccds in the camera produces gaps of @xmath13 to @xmath14 arcseconds on the sky .
the @xmath15 magnitudes and @xmath16 colors given in table 1 ( see next section ) , were taken from the catalog of godwin et al .
the recessional velocities are from the compilation by edwards et al .
these catalogs cover a region of 2.65 square degrees centred on the coma cluster and largely overlap our observed region . the compilation by edwards et al .
comprises 1174 galaxy redshifts of coma members , and foreground and background objects .
the mean velocity of the coma cluster is @xmath17 km s@xmath7 , and its velocity dispersion is @xmath18 km s@xmath7 .
our criterion for membership is the same as the one used by edwards et al . :
coma members are those objects with velocities in the range @xmath19 km s@xmath7 .
this range corresponds to a @xmath20 cut on the velocity distribution of the coma cluster galaxies . with this criterion ,
the number of coma members in the above velocity catalog is 745 .
the sample analyzed in this paper has been selected by magnitude , and includes only six galaxies with unknown redshifts ( see next sections ) , so incompleteness effects are negligible .
we select for our structural analysis ( see next section ) the 187 galaxies with @xmath21 mag .
to quantify the properties of each galactic structural component ( bulge and disc ) we use a parametric model to describe the observed radial profiles .
elliptical galaxies and the bulges of spiral galaxies are modeled with a srsic law ; discs are described by an exponential profile .
the effects of atmospheric blurring on srsic profiles ( trujillo et al . 2001b , 2001c ) have been taken into account using the algorithm described in trujillo et al .
( 2001a ; hereafter t01a ) and aguerri & trujillo ( 2002 ) . in the case of optical ground
based observations , the psf is dominated by atmospheric blurring and can be approximated by a gaussian or moffat function . for the observations presented here ,
we determined the psf from the radial profile of bright unsaturated stars in the image ; it was found to be well described by a moffat function with @xmath22 and fwhm=@xmath23 .
the srsic ( 1968 ) profile can be written as @xmath24.\ ] ] this law is a generalization of the de vaucouleurs ( 1959 ) profile and has been used widely in recent years as a good description of the profiles of dwarf and giant ellipticals and the bulges of spiral galaxies ( see , for example , caon , capaccioli , & donofrio 1993 ; andredakis , peletier , & balcells 1995 ; graham & colless 1997 ; balcells et al .
the parameters of the model are the central intensity @xmath25 , the effective radius @xmath26 , and the srsic index @xmath27 .
the quantity @xmath28 is defined so that the effective radius encloses half of the total luminosity ( i.e. , @xmath28 is the solution of @xmath29 , where @xmath30 and @xmath31 are the gamma function and the incomplete gamma function respectively ) .
the relation between the central intensity and the intensity at the effective radius is given by @xmath32 .
discs were modeled with an exponential profile , such that @xmath33 where the parameters are the scale length @xmath34 and the central intensity @xmath35 .
we assume projected elliptical symmetry for the bulge and disc , with the ellipticities ( @xmath36 and @xmath37 respectively ) , in general , different for each component .
the observed ellipticity at small radii is of course smaller than the true ellipticity because the seeing tends to make the isophotes rounder .
this effect is particularly important for the numerous dwarf galaxies and spiral galaxy bulges with effective radii as small as 2 - 3 arcseconds .
our algorithm takes this effect into account , and the intrinsic ellipticities for the bulge and the disc are determined simultaneously with the other structural parameters .
we tried to use the minimum number of components ( i.e. parameters ) , so we proceeded as follows .
every galaxy s major - axis surface brightness _ and _ ellipticity profile , generated using the iraf task ` ellipse ` , were simultaneously fitted with a two component model ( bulge plus disc ) .
the total flux of both components is computed by using the srsic analytical expressions extrapolated to infinity . for those objects in which we obtained a bulge - to - total luminosity ratio @xmath38
, we checked if it was possible to obtain a good fit ( i.e. , with a reduced chi - square value as small or smaller than in the two component case ) with just a bulge component .
when that was the case , we considered the object to be an elliptical galaxy using monte carlo simulations we have determined the uncertainty in the computed parameters .
details of how these simulations are constructed can be found in t01a and in aguerri & trujillo ( 2002 ) .
we created 150 artificial galaxies with structural parameters randomly distributed in the following intervals : * bulge only structures : 13 mag @xmath39 mag , 1@xmath40 , @xmath41 , and @xmath42 ( the lower limit of @xmath43 is due to the physical restrictions pointed out by trujillo et al.2001b ) .
* bulge+disc structures : @xmath44 mag @xmath39 mag , @xmath45 , @xmath46 , @xmath47 , @xmath48 , @xmath49 , and @xmath50 . a variety of starting parameters were used to ensure that our fits did not get trapped in local @xmath51 minima . from these simulations
we find that the bulge and disc parameters ( we include here the intrinsic ellipticity of each component ) can be determined with an accuracy of @xmath52% for galaxies with @xmath21 .
our structural analysis was therefore carried out only on the 187 galaxies brighter than @xmath53 mag ; the results are presented in table 1 . of these 187 galaxies ,
163 have recessional velocities between 4,000 and 10,000 km s@xmath7 ; for 32 of these , the parameters could not be measured with sufficient accuracy for various reasons , such as proximity to bright stars , uncertain deblending , bad columns of the detector , etc .
these galaxies are flagged with a negative sign preceding their @xmath54 ratio .
this group also includes galaxies which are irregular or peculiar .
also , because the structure of cd galaxies ( ngc 4874 and ngc 4889 ) is poorly understood , these galaxies will also be excluded .
the analysis presented in the following sections will be restricted to the 129 regular galaxies belonging to coma and having reliable parameters .
following the notation and taxonomy discussed in t01a , we have differentiated the galaxies according to the bulge - to - total luminosity ratio ( @xmath54 ) .
we will consider the value @xmath55 as the separation between early ( e / s0 ) and late ( s ) type galaxies . according to this criterion
, 61 galaxies were classified as late - type , and 68 galaxies as early - type . in the early - type group ,
14 objects show evidence of discs , while 54 are pure ellipticals . in the text
we will often refer to objects that have been modeled with only a bulge component as elliptical galaxies .
this does not necesarily imply some specific internal kinematics for these objects .
figure [ fig_bt2 ] shows a histogram of the @xmath54 values .
figure [ fig_dist_bt ] shows the cumulative distribution of early- and late - types as a function of the distance to the cluster center .
although the figure seems to indicate some evidence for the early - type galaxies to be distributed closer to the center than late - types , a kolmogorov - smirnov test shows that this is not statistically significant ( similar results were obtained considering the value @xmath56 as the separation between early- and late - type galaxies ) .
a more detailed analysis of this point is presented below . in figure [ gmp ]
we compare the @xmath1-band magnitudes quoted in the gmp catalog with the galaxy magnitudes derived from our ( sloan-@xmath10 passband ) structural parameters .
although the agreement is very good , there is a @xmath57 mag zero - point offset such that our estimates are fainter .
this may be due to an offset in the relative calibration or to the existence of a small color term between sloan-@xmath10 and the @xmath1-band magnitudes .
there is , however , no significant magnitude - dependent bias in our estimation of the magnitudes . although many studies in the past have been devoted to a morphological analysis of galaxies in the coma cluster ( see references in section 1 ) , the comparison between the structural parameters derived for individual galaxies from different papers is not straightforward and is often only of a limited nature .
some of the previous studies provide no tables of parameters but present only statistical results ( e.g. , gerbal et al .
1997 ) , or the analysis is purely qualitative , or they derive the structural parameters in a model independent way ( e.g. andreon 1996 ) . in our sample
we have 54 galaxies in common with dressler ( 1980 ) .
if we compare the morphological types derived there with those inferred from our @xmath54 ratio , we find good agreement . in a few cases , we classified objects as s0 or spiral galaxies that were classified as pure e by dressler
this is likely explained as a consequence of the different sensitivities of the two studies ( the faint discs we found in some of our objects were not detected by dressler ) . for only four galaxies classified by dressler as s0 , and as pure e by us , can the results be considered discrepant . however , this level of discrepancy ( @xmath58% ) is in agreement with the differences expected among classifications by different authors ( see lahav et al . 1995 ; andreon & davoust 1997 ) .
graham & guzmn ( 2003 ) have analysed a sample of 15 dwarf elliptical galaxies in the coma cluster imaged with the _ hubble space telescope _ , finding that all but two of them have a central point - like component . in principle , ignoring the presence of this central component could affect the determination of the structural parameters of bulges and discs ( balcells et al .
we have checked whether this could be the case for our sample .
we have seven objects in common with graham & guzmn ( 2003 ) .
although in our model we have not accounted for a central source , and we used major - axis light - profiles while graham & guzmn used geometric - mean light - profiles , the parameters obtained for the galaxies in common are in good agreement .
for instance , the typical deviation between estimates of the srsic indices is @xmath59 . only in one case ( gmp 3292 )
the two analyses give significantly different values .
therefore , at least for the dwarf ellipticals in coma , the presence of nuclear components may not have a large effect on our estimation of their structural parameters .
caon et al . ( 1993 ) and donofrio , capaccioli , & caon ( 1994 ) reported the existence of a correlation between the srsic index @xmath27 and the _ model - independent _ total luminosity of elliptical galaxiess in the virgo and fornax clusters .
this correlation has been shown to hold also for the bulges of spirals ( andredakis , peletier , & balcells 1995 ; graham 2001 ; balcells et al .
2003 ; macarthur , courteau , & holtzman 2003 ) , and also extends to the dwarf elliptical regime ( young & currie 1994 , 1995 ; binggeli & jerjen 1998 , graham & guzmn 2003 ) .
the correlation is such that more luminous bulges tend to have larger values of @xmath27 , their light distributions are more centrally concentrated . .
in this section we analyse this and other possible correlations existing among the structural parameters of the galaxies in our sample , and compare the results with previous studies .
the study by caon et al . was conducted in the @xmath15-band ; to make a proper comparison we have converted our surface brightness values obtained in the @xmath1-band using the @xmath16 color given in table 1 . figure [ fig_param1 ] shows the relation between the _ model - independent _ @xmath15-band magnitudes from godwin et al .
( 1983 ; we assume @xmath60 ) , and the three srsic parameters we obtained for the early - type galaxies in our sample .
we also plot the data for the galaxies analysed by caon et al . and donofrio et al .
in the virgo and fornax clusters .
the distributions of the structural parameters of elliptical galaxies in the three clusters are similar , although there are more low - luminosity dwarf ellipticals from coma delineating the lower arm of this forked distribution ( see graham & guzmn 2003 ) .
another way to show the correlations between the three srsic parameters is presented in figure [ fig_param2 ] .
galaxies from the three clusters exhibit similar values and relations between the parameters .
the correlation between the srsic index @xmath27 and luminosity ( figure [ fig_param1 ] ) is clear and holds for both the giants and dwarfs .
the rough correlation found between @xmath27 and @xmath61 is similar to the one found by caon et al .
( 1993 ) ( see also young & currie 1995 and graham et al .
for the spiral galaxies a similar correlation also exists between the bulge index @xmath27 and the bulge - to - total luminosity ratio ( andredakis et al .
1995 , graham et al . 2001 ; balcells et al .
trujillo et al . ( 2002b ) have interpreted this as a consequence of the relation between @xmath27 and the luminosity of the bulges .
the main difference between the galaxies in the three clusters is a group of bright galaxies in virgo which are absent in the core of coma and the fornax sample .
objects with similar magnitudes in coma are the two cd galaxies ngc 4874 and ngc 4889 , which have been excluded in this analysis . for the other two parameters ( surface brightness and effective radius )
one can distinguish two different regimes , with a transition region at @xmath62 close to the usual limit adopted to separate dwarf and ordinary ellipticals .
traditionally , they have been considered as two separate families of objects , although the exact separation in magnitude between each class is somewhat arbitrary
. edwards et al .
( 2002 ) considered dwarf galaxies as objects with @xmath63 mag .
this corresponds to m@xmath64 mag .
the point here is whether the relation between structural parameters and , therefore , the origin of giants and dwarfs is different and justifies this distinction .
graham & guzmn ( 2003 ) argued that the continuity between @xmath27 , and central bulge surface brightness , with luminosity demonstrates that dwarfs and ordinary ellipticals do not constitute two separate families of objects .
they explained the apparently different relations between @xmath65 and luminosity ( and @xmath65 and @xmath26 ) for the high- and low - luminosity ellipticals as an expected consequence of the above linear trends .
we have characterized the sizes of bulges and discs using the effective radius @xmath26 , and scale length @xmath34 , respectively .
figure [ fig_rerd ] shows the relative sizes of bulges and discs for our sample of 14 s0 and 61 spiral galaxies . from these plots
we see how the ratio @xmath66 is constant for galaxies with @xmath67 ( most of the spirals ) , with typical values in the range 0.150.30 . for galaxies with @xmath68 this
ratio increases to 0.6 and higher .
three early - type spiral galaxies have values @xmath69 these values are somewhat larger than those obtained for spiral galaxies in the field ( graham 2001 , 2003 ; macarthur et al .
2003 ) . to better understand the meaning of these results , we have compared the properties of the discs in the field with those in the core of coma .
disc galaxies brighter than @xmath70 mag are not present in our sample . on the other hand ,
graham s sample is not complete for galaxies fainter than @xmath71 mag , thus we use the common range @xmath72 mag @xmath73 mag .
the comparison is presented in figure [ fig_disc3 ] .
we obtained mean values of @xmath66=0.24 for coma and 0.17 for graham s sample .
a kolmogorov - smirnov test rejects the hypothesis that the two @xmath66 distributions are the same at @xmath74% .
applying student s t - test , the mean values of @xmath66 in both distributions are different at the 99.9 % level ( assuming the two distributions have unequal variances ) .
the larger average value of @xmath66 for galaxies in the denser environment of the coma cluster is compatible with the idea that the discs of spiral galaxies in the center of clusters are smaller than the discs of field galaxies with similar magnitudes and bulges .
in fact , if we assume that the sizes of the bulges are less affected by the enviroment than the size of the disc , we can estimate that the discs of the galaxies in the center of the cluster have 30% smaller scale - lengths .
aguerri et al .
( 2003 ) have found similar results from a study of the morphology of galaxies in a larger region of the coma cluster .
the reduction in the sizes of the discs is in agreement with what is expected in high density enviroments . in these enviroments
tidal forces play a crucial role truncating and heating the infalling disc galaxies ( moore et al . 1999 ,
gnedin 2003 ) .
this could also be explained if the suppression of star formation , as has been proposed by balogh , navarro , & morris ( 2000 ) , is more effective in the outer parts of the ( cluster ) spiral galaxies . as expected , due to the smaller values of @xmath34 but similar luminosities
, we find that in the @xmath75 plane ( figure [ fig_disc3 ] ) cluster disc galaxies tend to have brighter central surface brightnesses .
figure [ fig_edeb ] shows the seeing - corrected , projected ellipticities for the bulges and discs of the galaxies in our sample that have these two component .
the ellipticity for a thin disc is related to the inclination angle , @xmath76 , by : @xmath77 .
clearly the ellipticity of discs tends to be higher than those of bulges .
figure [ fig_elip ] shows the cumulative distributions of ellipticities for the different morphological types .
table [ elip ] shows the number of early- and late - type galaxies , together with the mean and standard deviation of the projected ellipticities .
the main results are that bulges of all morphological types have ellipticities with a similar distribution , and that the discs of s0s seem to have a slightly different distribution than the late - type galaxies ( a kolmogorov smirnov test rejects both distributions being the same at the 86 % confidence level ) .
however , it may simply be a result of small number statistics .
this needs to be investigated with a larger sample of s0 galaxies .
jorgensen & franx ( 1994 ) measured the ellipticities of a volume - limited sample of galaxies in the coma cluster .
however , they simply measured the global ellipticity of the galaxies without disentangling the disc and bulge components .
they concluded that s0 galaxies tend to have larger ellipticities than pure ellipticals , and argued that this indicates that some of the faint face - on s0 galaxies were misclassified as e types .
on the basis of our own results ( table [ elip ] ) , the differences in ellipticity found by those authors could indeed be due to the differences in ellipticity between the two structural components ( bulges and discs ) .
studying the velocity distributions of the different morphological types in coma allows one to test models of the origin and evolution of these types .
previous dynamical analyses of groups and clusters have been conducted by many teams .
for instance , zabludoff & franx ( 1993 ) have analyzed six rich clusters of galaxies ; schindler , binggeli , & bohringer ( 1999 ) and conselice , gallagher , & wyse ( 2001 ) have explored the virgo cluster ; held & mould ( 1997 ) and drinkwater , gregg , & colless ( 2001 ) studied the fornax cluster ; colless & dunn ( 1996 ) the coma cluster ; and cote et al .
( 1997 ) studied the centaurus a group .
in general , all these works show that late - type galaxies and dwarf ellipticals have broader velocity distributions as compared with giant ellipticals .
this has been interpreted as evidence that spiral and de galaxies are infalling into a virialized core dominated by giant ellipticals .
the coma cluster is not a simple relaxed cluster : a close examination of its central parts reveals the presence of two groups of galaxies ( fitchett & webster 1987 ; baier , fritze & tiersch 1990 ) dominated by the cd galaxies ngc 4874 and ngc 4889 .
the analysis of radial velocities by colless & dunn ( 1996 ) found these two concentrations to be dynamically different entities .
a third group , dominated by the galaxy ngc 4839 , is located 40@xmath78 to the sw of the cluster .
burns et al .
( 1994 ) claimed that the group had already been disrupted after its first passage through the cluster , while colless & dunn ( 1996 ) suggested that this group is falling into coma along the great wall .
the coma cluster is one the brightest extragalactic x - ray sources observed by _
( white et al . 1993 ) , _ asca _ ( watanabe et al . 1999 ) , and _ xmm _ ( briel et al . 2001 ) . in the x - rays images ,
the cluster appears elongated along the line connecting ngc 4874 and ngc 4889 , and numerous lumps and individual sources are visible .
based on these observations , neumann et al .
( 2001 ) discussed the morphology of the ngc 4839 group and concluded that the group is falling into coma for the first time , in agreement with the scenario proposed by colless & dunn ( 1996 ) .
gurzadyan & mazure ( 2001 ) analysed the substructure of the coma cluster using an s - tree method and also concluded that three subgroups exist .
furthermore , a study of the small scale structure conducted by conselice & gallagher ( 1998 ) discovered three additional aggregates , two of them equidistant between ngc 4874 and ngc 4889 , and the other near the giant elliptical ngc 4860 .
edwards et al . ( 2002 ) have conducted a detailed kinematical study of coma . analyzing differences in velocity between the giant and dwarf galaxies ( based on whether they are brighter or fainter than @xmath79 mag )
, they found that the giants follow a non - gaussian distribution in velocity , while the distribution of the dwarfs is compatible with a gaussian .
while we use the same compilation of redshifts , our quantitative morphological analysis allows us to study the motions of the galaxies as a function of their structural parameters .
the velocity distributions for ellipticals with @xmath80 ( dwarfs ) and @xmath3 ( giants ) are shown in fig .
[ fig_vel ] .
both distributions are symmetric with respect to the systemic velocity of the cluster .
the distribution for galaxies with @xmath81 is narrower and shows a larger number of galaxies with velocities close to the cluster systemic velocity .
this could indicate a concentration of galaxies with @xmath82 in the central parts of the cluster , in agreement with previous findings that dwarf galaxies are less concentrated than bright galaxies in coma ( quintana 1979 ) .
similar results have been found for the fornax cluster ( e.g. , caldwell 1987 ; drinkwater et al .
2001 ) , the cluster ac118 at @xmath83 ( andreon 2002 ) , and for a sample of other clusters between @xmath84 and @xmath85 ( adami et al .
table [ velo ] summarizes this analysis showing the mean velocity ( and dispersion ) obtained for the different types of galaxies discussed above . a kolmogorov
smirnov test applied to the velocity distributions of e galaxies with @xmath82 and @xmath80 shows some evidence ( 70% of confidence level ) for the two distributions being different .
table [ velo ] shows that the ratio between cluster velocity dispersion for these galaxies is @xmath86 .
edwards et al . also found a slightly larger dispersion for dwarf than for giant galaxies ( 1096@xmath87 45 and @xmath88 30 km s@xmath7 respectively ) in coma .
if the central 0.28 square degrees of the coma cluster were fully relaxed , by dynamical relaxation ( binney & tremaine , 1987 ) , this should correspond to mass ratios between the two groups of galaxies @xmath89 . assuming a similar mass - to - light ratio for all the ellipticals
, this would imply a mean difference in magnitude between both groups @xmath90 , which is smaller than the actual difference ( 2.4 mag ) .
for instance , edwards et al . have performed monte carlo simulation of the velocities in a cluster fully relaxed , finding that the ratio between the velocity dispersion for dwarfs and giants should be @xmath4 .
edwards et al . argued that the differences in velocity between the two groups of ellipticals are the result of the merging of subclusters which were partially relaxed . in the particular case of the coma cluster , this could be supported also by the large scale distribution of the groups associated to ngc 4874 and ngc 4889 respectively . in the compilation of redshifts by edwards
et al . (
2002 ) there are 745 coma members ; the rest are 421 background and 15 foreground objects .
figure [ fig_histo ] shows a histogram with the distribution of redshifts in this catalog .
the diagram on the left is dominated by a big bump at @xmath17 km s@xmath7 , which corresponds to the coma cluster .
there are two other concentrations at higher redshift , peaked at @xmath91 and @xmath8 km s@xmath7 ; these can be seen in more detail in the diagram on the right of figure [ fig_histo ] .
the contrast in galaxy density between these structures and the average density distribution is @xmath4 .
figure [ fig_histo ] also shows a low density region from 10,000 km s@xmath7 to 20,000 km s@xmath7 that has been discussed in lindner et al .
( 1995 ) .
as the limiting magnitude of the sample with measured redshifts is unclear , it is not possible to make a comparison with the values reported by arnouts et al .
( 1997 ) for the expected density of galaxies in the field .
the range of velocities in both of the high - density structures beyond coma are too high for those expected from a typical cluster ; however , both structures show evidence of substructure within them .
the closer concentration can be split into two substructures , one at @xmath6 and the other at @xmath91 km s@xmath7 , both having velocity dispersion @xmath92 km s@xmath7 .
although it is difficult to appreciate in the figure , we tentatively identify in the more distant concentration of galaxies two overlapping structures at @xmath93 and @xmath94 km s@xmath7 , with dispersions of @xmath95 and @xmath18 km s@xmath7 respectively .
it is interesting to note that both structures seem to extend over the full spatial region ( @xmath96 square degrees ) of the velocity catalog without a clear spatial concentration ; they subtend angles much larger than those expected for a cluster at these redshifts . for instance , a typical cluster with size of 1 mpc would subtend an angle of @xmath97 arcminutes at @xmath98 km s@xmath7 .
the three dimensional structure of the background galaxies therefore resembles large , low density regions ( voids ) separated by thin walls .
1 . we have determined the morphological types and structural parameters of 187 galaxies in the central part of the coma cluster ( 0.28 square degrees ) .
the analysis extends down to @xmath99 mag .
the results of our quantitative - based ( @xmath54 ratio ) galaxy type determinations are in good agreement with previous subjective classifications
we have shown various correlations between the surface brightness , effective radius , luminosity , and the srsic shape index @xmath27 for ellipticals and the bulges of spirals .
all results are in agreement with past studies of elliptical galaxies in the virgo and the fornax cluster . the strongest correlation that between the srsic index , @xmath27 , and luminosity .
3 . the bulges of all morphological types show a similar distribution of projected ellipticities , while there is marginal evidence that the discs of s0 and spirals have different projected - ellipticity distributions .
we have not found bright spiral galaxies ( @xmath100 mag ) in the core of coma . in the magnitude interval @xmath101 , the scale length of the discs of coma spiral galaxies are 30% smaller than for field galaxies . this may be evidence of environment - driven evolution .
we thank a. aparicio and a. marn who kindly provided us with the images of coma .
we also thanks to m. colless who has provided us with the compilation of redshifts .
the isaac newton telescope ( int ) is operated on the island of la palma by the isaac newton group and the instituto de astrofsica de canarias ( iac ) in the spanish observatory roque de los muchachos of the iac .
adami , c. , mazure , a. , ulmer , m. p. , & savine , c. 2001 , a&a , 371 , 11 aguerri , j. a. l. , & trujillo , i. , 2002 , mnras , 333 , 633 aguerri , j. a. l. , iglesias - pramo , j. , vlchez , j. m. & muoz - tun , c. , 2003 , in preparation andredakis , y. c. , peletier , r. f. , & balcells , m. 1995 , mnras , 275 , 874 1995 andreon , s. 1996 , a&a , 314 , 763 andreon , s. , davoust , e. , & poulain , t. 1997 , a&as , 126 , 67 andreon , s. , davoust , e. , michard , r. , nieto , j .- l . , & poulain , t. 1995 , a&as , 116 , 429 andreon , s. , & davoust , e. 1997 , a&a , 319 , 747 arnouts , s. , de lapparent , v. , mathez , g. , mazure , a. , mellier , y. , bertin , e. & kruszewski , a. 1997 baier , f. w. , fritze , k. , & tiersch , h. 1990 , an , 311 , 89 balcells , m. , graham , a. w. , domnguez - palmero , l. , & peletier , r. f. 2003 , apj , 582l , 79 balogh , m. l. , navarro , j. f. , & morris , s. l. 2000 , apj , 540 , 113 beijersbergen , m. , hoekstra , h. , van dokkum , p. g. , & van der hulst , t. 2002 , mnras , 329 , 385 bernstein , g. m. , nichol , r. c. , tyson , j. a. , ulmer , m. p. , & wittmann , d. 1995 , aj , 110 , 1507 binggeli , b. , & jerjen , h. 1998 , a&a , 333 , 17 binney , j. & tremaine , s. 1987 , galactic dynamics ( princeton : princeton univ . press ) briel , u. g. 2001 et al .
2001 , a&a , 365 , l60 burns , j. o. , roettiger , k. , ledlow , m. , & klypin , a. 1994 , apj , 427 , l87 caldwell , n. 1987 , aj , 94 , 1116 caon , n. , capaccioli , m. , & donofrio , m. 1993 , mnras , 265 , 1013 castander , f. j. et al .
2001 , aj , 121 , 5 cellone , s. a. , forte , j. c. , & geisler , d. 1994 , ajss , 93 , 397 colless , m. , & dunn , a. m. 1996 , apj , 458 , 435 conselice , c. j. , & gallagher , j. s. iii 1998 , mnras , 297 , 34 conselice , c. j. , gallagher , j. s. iii , & wyse , r. f. g. 2001 , apj , 559 , 791 cote , s. , freeman , k. c. , carignan , c. , & quinn , p. j. 1997 , aj , 114 , 1313 de jong , r. s. 1996 , a&as , 118 , 557 donofrio , m. , capaccioli , m. , & caon , n. 1994 , mnras , 271 , 523 dressler , a. 1980 , apjss , 42 , 565 drinkwater , m. j. , gregg , m. j. , & colless , m. 2001 , apj , 548 , l39 edwards , s. a. , colless , m. , bridges , t. j. , carter , d. , mobasher , b. , & poggianti , b. m. 2002 , apj , 567 , 178 fitchett , m. , & webster , r. 1987 , apj , 317 , 653 gallagher , j. s. , conselice , c. j. , & wise , r. f. g. 2001 , in dwarf galaxies and their environment .
edited by klaas s. de boer , ralf - juergen dettmar , and uli klein .
isbn 3826592646 , p. 213 gavazzi , g. , zibetti , s. , boselli , a. , franzetti , p. , scodeggio , m. , & martocchi , s. 2001 , a&a , 372 , 29 gerbal , d. , lima - neto , g. b. , marquez , i. , & verhagen , h. 1997 , mnras , 285 , l41 gnedin , o. , 2003 , apj , in press godwin , j. g. , metcalfe , n. , & peach , j. v. 1983 , mnras , 202 , 113 graham , a. w. 2001 , aj , 121 , 820 graham , a. w. 2003 , aj , 125 , 3398 graham , a.w . , & colless , m.m .
1997 , mnras , 287 , 221 graham , a. , guzmn , r. 2003 , aj , 125 , 2936 graham , a. , lauer , t. r. , colless , m. , & postman , m. 1996 , apj , 465 , 534 gurzadyan , v. g. , & mazure , a. 2001 , new astronomy , 6 , 43 held , v. e. , & mould , j. r. 1994 , aj , 107 , 1307 jorgensen , i. , & franx , m. 1994 , apj , 433 , 553 karachentsev , i. d. , karachentsev , v. e. , richter , g. m. , & vennik , j. a. 1995 , a&a , 296 , 643 kashikawa , n. et al .
1998 , apj , 500 , 750 kashikawa , n. , shimasaku , k. , yagi , m. , yasuda , n. , doi , m. , & okamura , s. 1995 , apj , 452 , l99 khosroshahi , h. g. , wadadekar , y. , kembhavi , a. , & mobasher , b. 2000 , apj , 531 , l103 komiyama , y. et al .
2002 , apjs , 138 , 265 lahav , o. , et al .
1995 , sci , 267 , 859 lindner , u. , einasto , j. , einasto , m. , freudling , w. , fricke , k. , & tago , e. 1995 , a&a , 301 , 329 lobo , c. , biviano , a. , durret , f. , gerbal , d. , le fevre , o. , mazure , a. , & slezak , e. 1997 , a&ass , 122 , 409 lucey , j. r. , guzmn , r. , carter , d. , & terlevich , r. j. 1991 , mnras , 253 , 584 macarthur , l.,a .
courteau , s. , & holtzman , j. a. 2003 , apj , 582 , 689 marn - franch , a. , & aparicio , a. 2002 , apj , 568 , l74 mehlert , d. , saglia , r. p > , bender , r. , & wegner , g. 2000 , a&as , 141 , 449 mobasher , b. et al .
2001 , apjs , 137 , 279 mobasher , b. , guzmn , r. , aragon - salamanca , a. , & zepf , s. 1999 , mnras , 304 , 225 moore , b. , lake , g. , quinn , t. , & stadel , j. , 1999 , mnras , 304 , 465 neumann , d. m. et al .
2001 , aa , 365 , l74 pahre , m. 1999 , apjss , 124 , 127 quintana , h. 1979 , aj , 84 , 15 rood , h. j. , & baum , w. a. 1967 , aj , 72 , 398 schindler , s. , binggeli , b. , & bohringer , h. 1999 , a&a , 343 , 420 secker , j. harris , w. e. 1997 , pasp , 109 , 1364 srsic , j. l. 1968 , atlas de galaxias australes , crdoba , argentina : observatorio astronmico terlevich , a. i. , caldwell , n. , & bower , r. g. 2001 , mnras , 326 , 1547 trentham , n. 1998 , mnras , 293 , 71 trujillo , i. , aguerri , j. a. l. , gutirrez , c. m. , & cepa , j. 2001a , aj , 122 , 38 ( t01a ) trujillo , i. , aguerri , j. a. l. , cepa , j. , & gutirrez , c. m. 2001c , mnras , 328 , 977 trujillo , i. , aguerri , j. a. l. , cepa , j. , & gutirrez , c. m. 2001b , mnras , 321 , 269 trujillo , i. , aguerri , j. a. l. , gutirrez , c. m. , caon , n. & cepa , j. 2002a , apj , 573 , l9 ( t02a ) trujillo , i. , asensio ramos , a. , rubino - martin , j. a. , graham , a. w. , aguerri , j. a. l. , cepa , j. , & gutierrez , c. m. 2002b , mnras , 333 , 510 white , s. , briel , u. g. , & henry , j. p. 1993 , mnras , 261 , l8 watanabe , m. , yamashita , k. , furuzawa , a. , kunieda , h. , tawara , y. , & honda , h. 1999 , apj , 527 , 80 young , c. k. , & currie , m. j. 1994 , mnras , 268 , l11 young , c. k. , & currie , m. j. 1995 , mnras , 273 , 1141 zabludoff , a. i. , & franx , m. 1993 , aj , 106 , 1314 + gmp & & b & b - r & v & x & y & @xmath102 & @xmath61 & n & @xmath103 & @xmath104 & h & @xmath105 & b / t + + + 2347 & d098 & 15.85 & 1.91 & 6848 & 16.90 & -4.27 & 18.04 & 0.70 & 1.29 & 0.11 & 18.82 & 4.22 & 0.47 & 0.17 + 2374 & n4911 d082 & 13.91 & 1.67 & 7987 & 16.15 & -10.85 & 18.63 & 1.10 & 2.24 & 0.08 & 19.37 & 9.25 & 0.44 & -0.11 + 2376 & & 18.23 & 1.77 & 6049 & 16.15 & -4.36 & 23.16 & 6.70 & 1.83 & 0.45 & & & & 1.00 + 2385 & & 17.62 & 1.82 & 7040 & 5.90 & -7.74 & 20.37 & 2.15 & 2.19 & 0.25 & & & & 1.00 + 2390 & ic4051 d143 & 14.47 & 1.82 & 4968 & 5.82 & 2.19 & 21.79 & 15.80 & 3.86 & 0.36 & & & & 1.00 + 2393 & d062 & 16.51 & 1.90 & 8302 & 5.75 & -11.25 & 20.99 & 3.28 & 4.56 & 0.27 & 21.04 & 7.53 & 0.74 & 0.68 + 2399 & & 18.78 & 9.99 & 5837 & 5.68 & -0.15 & 22.09 & 2.95 & 1.35 & 0.32 & & & & 1.00 + 2408 & & 18.49 & 1.73 & & 5.46 & -8.72 & 22.97 & 3.15 & 0.83 & 0.45 & 23.71 & 7.90 & 0.61 & 0.43 + 2417 & n4908 d167 & 14.91 & 1.87 & 8742 & 5.17 & 4.34 & 20.34 & 6.96 & 3.81 & 0.34 & & & & 1.00 + 2421 & & 17.98 & 1.90 & 8150 & 5.10 & -13.68 & 22.01 & 1.83 & 1.17 & 0.25 & 20.96 & 4.18 & 0.53 & 0.19 + 2440 & ic4045 d168 & 15.17 & 1.85 & 6899 & 4.54 & 7.18 & 21.05 & 1.83 & 1.68 & 0.30 & & & & -1.00 + 2441 & n4907 d205 & 14.65 & 1.74 & 5773 & 4.51 & 11.23 & 19.71 & 2.01 & 3.20 & 0.11 & 19.35 & 8.86 & 0.06 & -0.10 + 2457 & d117 & 16.56 & 1.88 & 8604 & 4.27 & -2.94 & 18.48 & 0.66 & 1.28 & 0.29 & 18.95 & 3.10 & 0.46 & -0.16 + 2478 & & 18.09 & 1.86 & 8752 & 3.84 & -8.13 & 21.56 & 3.43 & 1.35 & 0.41 & & & & 1.00 + 2489 & d191 & 16.69 & 1.77 & 6582 & 3.65 & 7.78 & 18.02 & 0.66 & 0.79 & 0.28 & 18.48 & 2.74 & 0.55 & 0.20 + 2510 & d116 & 16.13 & 1.90 & 8353 & 3.24 & -0.47 & 19.11 & 1.44 & 2.31 & 0.15 & 19.92 & 4.51 & 0.30 & -0.41 + 2516 & ic4042 d144 & 15.34 & 1.86 & 6354 & 3.19 & 0.03 & 19.84 & 3.68 & 3.62 & 0.16 & 23.19 & 51.89 & 0.18 & 0.28 + 2519 & & 18.68 & 1.72 & 6216 & 3.21 & 8.72 & 22.15 & 1.24 & 1.76 & 0.21 & 22.44 & 5.55 & 0.21 & 0.14 + 2529 & & 18.63 & 1.87 & 8826 & 2.89 & 4.45 & 21.05 & 1.83 & 1.68 & 0.10 & & & & 1.00 + 2531 & & 17.63 & 1.70 & 18442 & 2.89 & -4.55 & 22.77 & 0.96 & 11.43 & 0.31 & 20.14 & 3.53 & 0.23 & -0.03 + 2535 & ic4041 d145 & 15.93 & 1.90 & 7059 & 2.82 & 1.54 & 19.32 & 1.65 & 1.87 & 0.20 & 19.39 & 4.61 & 0.49 & -0.35 + 2541 & n4906 d118 & 15.44 & 1.98 & 7497 & 2.54 & -2.82 & 19.55 & 2.59 & 2.84 & 0.20 & 19.86 & 4.58 & 0.04 & 0.52 + 2559 & ic4040 d169 & 15.44 & 9.99 & 7801 & 2.08 & 5.25 & 23.00 & 3.49 & 2.41 & 0.12 & 18.68 & 4.72 & 0.83 & 0.13 + 2584 & d192 & 16.14 & 1.79 & 5465 & 1.63 & 10.52 & 20.28 & 1.10 & 5.90 & 0.05 & 18.63 & 4.91 & 0.68 & 0.12 + 2585 & & 18.44 & 1.73 & 6914 & 1.62 & -1.69 & 22.31 & 4.12 & 1.82 & 0.41 & & & & 1.00 + 2591 & d051 & 18.50 & 1.86 & 8356 & 1.41 & -2.17 & 22.94 & 4.11 & 2.13 & 0.09 & & & & 1.00 + 2603 & & 17.36 & 1.80 & 8152 & 1.17 & -8.81 & 21.60 & 4.74 & 2.15 & 0.50 & & & & 1.00 + 2615 & d063 & 16.97 & 1.90 & 6679 & 0.99 & -12.29 & 19.30 & 0.90 & 1.02 & 0.24 & 19.74 & 3.46 & 0.33 & -0.18 + 2619 & & 18.38 & 1.71 & 6279 & 0.97 & 17.10 & 22.16 & 1.19 & 1.04 & 0.27 & 21.21 & 3.85 & 0.24 & 0.07 + 2633 & & 18.48 & 1.75 & -359 & 0.48 & 11.37 & 20.78 & 0.63 & 2.38 & 0.37 & 20.37 & 2.30 & 0.34 & 0.12 + 2651 & d147 & 16.19 & 1.85 & 7700 & 0.07 & 0.08 & 19.44 & 1.42 & 2.33 & 0.39 & 19.82 & 5.53 & 0.54 & -0.26 + 2654 & d119 & 16.38 & 1.90 & 6984 & 9.98 & -0.90 & 17.90 & 0.73 & 1.79 & 0.08 & 19.86 & 4.12 & 0.37 & 0.41 + 2692 & & 18.20 & 1.78 & 7962 & 9.28 & -2.66 & 23.76 & 4.94 & 3.55 & 0.14 & 20.89 & 3.66 & 0.66 & 0.52 + 2711 & & 18.90 & 9.99 & 19792 & 8.90 & 14.20 & 19.49 & 0.60 & 1.96 & 0.21 & & & & 1.00 + 2727 & ic4026 d170 & 15.73 & 1.77 & 8161 & 8.69 & 4.56 & 18.52 & 1.10 & 1.76 & 0.15 & 19.43 & 4.20 & 0.20 & -0.29 + 2728 & & 18.15 & 1.73 & 7965 & 8.70 & 16.57 & 20.21 & 0.38 & 1.52 & 0.10 & 19.53 & 2.69 & 0.56 & 0.05 + 2736 & & 18.21 & 1.76 & 4869 & 8.59 & -4.34 & 19.84 & 0.63 & 0.62 & 0.28 & 20.37 & 2.19 & 0.25 & 0.17 + 2753 & & 18.10 & 1.86 & 7767 & 8.26 & -7.65 & 21.40 & 2.33 & 0.90 & 0.22 & 22.06 & 4.45 & 0.23 & 0.48 + 2777 & & 18.21 & 1.71 & 6202 & 7.97 & 2.30 & 19.17 & 0.74 & 2.64 & 0.06 & & & & -1.00 + 2778 & & 16.69 & 1.81 & 5122 & 7.95 & -2.03 & 20.43 & 1.08 & 1.25 & 0.38 & 19.27 & 2.86 & 0.05 & -0.06 + 2783 & & 17.37 & 1.83 & 5318 & 7.90 & -9.33 & 23.13 & 8.44 & 2.64 & 0.08 & & & & -1.00 + 2784 & & 18.36 & 1.81 & 7838 & 7.90 & 7.57 & 21.38 & 1.64 & 0.97 & 0.13 & 22.05 & 4.54 & 0.19 & 0.33 + 2787 & & 18.46 & 1.65 & 9873 & 7.85 & 5.30 & 22.46 & 3.79 & 1.71 & 0.15 & & & & 1.00 + 2795 & n4895 d206 & 14.38 & 9.99 & 8469 & 7.76 & 13.89 & 20.19 & 8.84 & 2.49 & 0.10 & 21.20 & 23.80 & 0.69 & 0.74 + 2799 & & 18.70 & 9.99 & 6136 & 7.70 & 1.00 & 21.27 & 1.39 & 1.73 & 0.07 & & & & -1.00 + 2800 & & 18.38 & 9.99 & 7015 & 7.70 & -11.19 & 22.77 & 1.78 & 1.80 & 0.05 & 22.37 & 5.19 & 0.13 & 0.18 + 2815 & n4894 d122 & 15.87 & 1.74 & 4636 & 7.45 & -0.20 & 17.96 & 0.73 & 1.52 & 0.12 & 19.15 & 4.25 & 0.60 & 0.31 + 2839 & ic4021 d172 & 16.01 & 1.75 & 5724 & 7.06 & 4.22 & 17.37 & 0.69 & 1.56 & 0.22 & 18.94 & 2.73 & 0.10 & 0.35 + 2852 & & 17.80 & 1.79 & 7405 & 6.81 & -6.22 & 21.00 & 0.88 & 1.49 & 0.29 & 20.78 & 3.04 & 0.24 & 0.13 + 2856 & & 18.23 & 1.57 & 8134 & 6.77 & 4.94 & 23.83 & 11.98 & 1.29 & 0.72 & 18.96 & 1.63 & 0.92 & 0.82 + 2861 & d173 & 16.26 & 1.85 & 7511 & 6.65 & 6.27 & 18.88 & 1.42 & 2.20 & 0.12 & 20.21 & 4.81 & 0.28 & 0.49 + 2866 & d064 & 16.90 & 1.79 & 7030 & 6.59 & -11.34 & 20.69 & 1.84 & 2.61 & 0.11 & 21.49 & 4.71 & 0.05 & 0.47 + 2895 & & 18.64 & 2.04 & 26236 & 6.05 & -0.77 & 21.78 & 0.69 & 0.83 & 0.28 & 20.17 & 2.10 & 0.56 & 0.06 + 2897 & d099 & 16.98 & 1.53 & 9889 & 6.00 & -6.42 & 21.67 & 2.03 & 5.37 & 0.64 & 19.89 & 2.48 & 0.08 & 0.17 + 2910 & d100 & 16.25 & 1.41 & 5136 & 5.78 & -6.28 & 20.58 & 3.54 & 1.82 & 0.30 & & & & -1.00 + 2912 & n4895a d207 & 16.07 & 1.80 & 6752 & 5.82 & 11.97 & 20.69 & 5.13 & 3.58 & 0.35 & & & & 1.00 + 2914 & & 17.18 & 1.81 & 7436 & 5.74 & 11.34 & 20.27 & 0.97 & 1.35 & 0.06 & 20.55 & 4.03 & 0.14 & 0.15 + 2921 & n4889 d148 & 12.62 & 1.91 & 6508 & 5.57 & 0.35 & 19.17 & 2.62 & 1.88 & 0.34 & & & & 1.00 + 2922 & ic4012 d174 & 15.93 & 1.86 & 7207 & 5.57 & 6.46 & 19.60 & 3.08 & 3.26 & 0.20 & & & & 1.00 + 2923 & & 17.65 & 9.99 & 8664 & 5.57 & -11.86 & 21.40 & 2.87 & 0.97 & 0.43 & 22.17 & 7.12 & 0.62 & 0.48 + 2929 & & 18.66 & 1.90 & 6266 & 5.45 & -0.77 & 22.94 & 3.49 & 1.73 & 0.05 & & & & 1.00 + 2940 & ic4011 d150 & 16.08 & 1.82 & 7244 & 5.21 & 2.00 & 17.21 & 0.42 & 1.32 & 0.06 & 18.78 & 2.31 & 0.09 & 0.24 + 2943 & & 17.66 & 9.99 & 13298 & 5.18 & 16.84 & 24.98 & 0.81 & 67.60 & 0.17 & 21.45 & 5.48 & 0.12 & 0.01 + 2945 & d065 & 16.15 & 1.77 & 6150 & 5.14 & -11.73 & 19.27 & 1.20 & 1.46 & 0.17 & 19.74 & 4.73 & 0.55 & 0.29 + 2956 & d084 & 16.20 & 1.98 & 6566 & 5.02 & -9.79 & 22.74 & 12.49 & 4.67 & 0.10 & 19.63 & 3.61 & 0.63 & 0.86 + 2960 & & 16.78 & 1.74 & 5827 & 5.00 & 3.22 & 20.03 & 0.99 & 2.05 & 0.14 & 19.73 & 3.78 & 0.63 & 0.24 + 2975 & n4886 d151 & 14.83 & 1.76 & 6363 & 4.78 & 1.01 & 21.39 & 16.37 & 7.00 & 0.05 & & & & 1.00 + 2976 & & 18.14 & 1.94 & 6693 & 4.75 & 11.09 & 23.06 & 6.04 & 1.36 & 0.45 & & & & -1.00 + 2985 & & 17.87 & 1.65 & 5317 & 4.63 & -0.37 & 21.96 & 3.25 & 1.65 & 0.33 & & & & 1.00 + 2989 & & 17.05 & 9.99 & 7675 & 4.48 & 16.17 & 18.42 & 0.57 & 1.70 & 0.18 & 20.75 & 3.81 & 0.29 & 0.35 + 3012 & & 17.49 & 1.83 & 8033 & 4.13 & -14.41 & 22.29 & 2.20 & 1.65 & 0.14 & 21.18 & 5.41 & 0.51 & 0.20 + 3017 & & 17.91 & 1.65 & 6965 & 4.01 & -1.52 & 23.05 & 4.53 & 3.92 & 0.15 & & & & 1.00 + 3034 & & 18.06 & 1.70 & 5994 & 3.69 & -1.82 & 23.72 & 6.90 & 2.72 & 0.10 & & & & 1.00 + 3055 & n4881 d217 & 14.73 & 1.87 & 6757 & 3.32 & 16.56 & 21.74 & 12.55 & 4.44 & 0.05 & & & & 1.00 + 3068 & d123 & 16.47 & 1.93 & 7716 & 3.07 & -2.43 & 19.22 & 0.80 & 1.83 & 0.29 & 19.44 & 3.89 & 0.45 & -0.14 + 3071 & & 17.17 & 1.18 & 8909 & 2.95 & -13.46 & 21.56 & 4.40 & 2.22 & 0.10 & 21.60 & 0.55 & 0.69 & 1.00 + 3073 & n4883 d175 & 15.43 & 1.89 & 8130 & 2.93 & 3.83 & 18.45 & 1.24 & 1.45 & 0.13 & 18.98 & 4.16 & 0.25 & 0.27 + 3084 & d193 & 16.43 & 1.82 & 7538 & 2.73 & 9.45 & 17.59 & 0.51 & 0.75 & 0.13 & 18.76 & 2.25 & 0.15 & -0.21 + 3092 & d085 & 17.55 & 1.59 & 8094 & 2.66 & -10.49 & 19.95 & 0.79 & 2.15 & 0.04 & 19.90 & 1.87 & 0.12 & 0.33 + 3098 & & 18.63 & 1.74 & 6618 & 2.46 & -0.02 & 22.57 & 3.06 & 1.62 & 0.09 & & & & 1.00 + 3113 & & 17.82 & 1.81 & 7601 & 1.99 & 7.66 & 21.54 & 3.26 & 1.88 & 0.24 & & & & 1.00 + 3121 & & 17.34 & 1.76 & 7418 & 1.93 & 6.16 & 20.38 & 0.55 & 0.92 & 0.07 & 20.36 & 3.12 & 0.10 & 0.06 + 3126 & & 17.55 & 1.82 & 7906 & 1.81 & -8.27 & 20.17 & 0.52 & 3.12 & 0.61 & 19.46 & 2.73 & 0.60 & 0.06 + 3129 & & 17.94 & 1.71 & 6815 & 1.67 & 10.41 & 21.81 & 4.48 & 2.22 & 0.54 & & & & 1.00 + 3133 & & 17.23 & 1.92 & 9770 & 1.62 & -2.76 & 19.30 & 1.01 & 0.80 & 0.24 & 19.87 & 2.57 & 0.27 & 0.32 + 3146 & & 18.54 & 1.58 & 5312 & 1.29 & 0.72 & 22.80 & 3.22 & 1.12 & 0.09 & & & & 1.00 + 3154 & & 18.64 & 1.68 & & 1.12 & 7.63 & 21.60 & 2.11 & 1.31 & 0.19 & & & & 1.00 + 3160 & & 18.88 & 9.99 & 9546 & 0.98 & 5.03 & 23.32 & 1.15 & 1.26 & 0.03 & 22.34 & 4.69 & 0.56 & 0.10 + 3166 & & 18.37 & 1.79 & 8443 & 0.92 & 1.27 & 22.73 & 3.75 & 1.98 & 0.22 & & & & 1.00 + 3170 & ic3998 d152 & 15.70 & 1.90 & 9399 & 0.88 & 0.19 & 19.33 & 1.62 & 3.09 & 0.27 & 19.71 & 4.89 & 0.25 & -0.32 + 3172 & & 18.69 & 1.83 & 27401 & 0.86 & 18.25 & 21.35 & 1.67 & 4.86 & 0.11 & & & & 1.00 + 3178 & d101 & 16.18 & 1.84 & 7991 & 0.73 & -6.81 & 20.97 & 3.76 & 4.69 & 0.10 & 20.38 & 4.25 & 0.66 & 0.82 + 3192 & & 18.17 & 1.49 & 5959 & 0.48 & -14.02 & 21.67 & 0.83 & 1.28 & 0.01 & 21.80 & 4.11 & 0.15 & 0.10 + 3196 & & 18.35 & 1.82 & 6847 & 0.41 & -4.86 & 21.91 & 2.40 & 2.02 & 0.18 & & & & 1.00 + 3204 & & 18.54 & 1.88 & 8340 & 0.32 & 12.33 & 21.02 & 1.71 & 1.87 & 0.11 & & & & 1.00 + 3205 & & 17.61 & 1.83 & 6237 & 0.31 & -6.19 & 21.56 & 3.59 & 1.62 & 0.30 & & & & 1.00 + 3206 & d126 & 16.36 & 1.79 & 6869 & 0.31 & -0.74 & 19.88 & 0.90 & 3.64 & 0.14 & 18.80 & 2.67 & 0.45 & 0.18 + 3213 & d153 & 16.14 & 1.83 & 6752 & 0.21 & 1.43 & 20.57 & 3.32 & 4.14 & 0.06 & & & & 1.00 + 3222 & d125 & 16.47 & 1.75 & 6925 & -0.11 & -2.76 & 19.53 & 1.89 & 4.77 & 0.12 & & & & 1.00 + 3225 & & 18.33 & 1.70 & 16911 & -0.13 & -6.71 & 19.97 & 0.82 & 0.95 & 0.20 & 20.31 & 2.59 & 0.60 & 0.34 + 3245 & & 18.42 & 2.18 & 29670 & -0.42 & 12.87 & 20.55 & 1.68 & 3.20 & 0.33 & 24.44 & 15.61 & 0.20 & 0.53 + 3248 & & 18.75 & 1.78 & 7519 & -0.43 & 10.43 & 22.38 & 3.09 & 1.85 & 0.11 & & & & 1.00 + 3254 & d127 & 16.57 & 1.84 & 7549 & -0.56 & -0.15 & 19.56 & 1.09 & 3.41 & 0.26 & 19.97 & 3.14 & 0.45 & 0.44 + 3258 & & 18.61 & 1.68 & 32977 & -0.56 & 12.52 & 18.01 & 0.36 & 1.87 & 0.83 & 19.51 & 1.37 & 0.34 & 0.15 + 3262 & d102 & 16.77 & 1.82 & 3690 & -0.64 & -6.96 & 21.79 & 5.25 & 3.55 & 0.32 & & & & -1.00 + 3269 & d128 & 16.12 & 1.75 & 7860 & -0.69 & -1.01 & 18.07 & 0.60 & 1.79 & 0.28 & 18.95 & 2.77 & 0.57 & 0.31 + 3291 & d154 & 16.41 & 1.78 & 6927 & -0.98 & 0.99 & 22.03 & 4.19 & 3.62 & 0.26 & 20.80 & 4.37 & 0.39 & 0.55 + 3292 & & 17.70 & 1.85 & 4890 & -1.05 & 1.82 & 19.90 & 0.47 & 1.17 & 0.06 & 19.94 & 2.02 & 0.18 & 0.12 + 3296 & n4875 d104 & 15.88 & 1.96 & 8005 & -1.08 & -3.80 & 18.55 & 1.12 & 2.32 & 0.13 & 19.34 & 3.21 & 0.21 & -0.43 + 3298 & & 17.26 & 1.79 & 6833 & -1.11 & -11.63 & 21.30 & 1.60 & 1.48 & 0.41 & 21.19 & 6.86 & 0.55 & 0.13 + 3302 & & 17.33 & 1.68 & 5617 & -1.13 & 11.71 & 20.85 & 2.56 & 2.11 & 0.24 & & & & 1.00 + 3312 & & 18.68 & 1.78 & 7293 & -1.27 & 2.88 & 20.92 & 1.29 & 1.73 & 0.09 & & & & -1.00 + 3313 & & 17.53 & 1.83 & 6262 & -1.29 & -8.69 & 19.70 & 1.30 & 2.07 & 0.07 & & & & -1.00 + 3329 & n4874 d129 & 12.78 & 9.99 & 7166 & -1.56 & -0.68 & 19.86 & 8.02 & 1.25 & 0.15 & & & & 1.00 + 3336 & & 18.47 & 9.99 & 13278 & -1.57 & -3.88 & 21.49 & 1.60 & 2.15 & 0.14 & & & & 1.00 + 3339 & & 17.54 & 1.78 & 6279 & -1.66 & -6.43 & 20.40 & 1.69 & 1.07 & 0.11 & 22.49 & 5.14 & 0.17 & 0.61 + 3340 & & 18.54 & 1.93 & 4193 & -1.68 & -2.16 & 22.22 & 2.46 & 1.67 & 0.19 & & & & 1.00 + 3367 & n4873 d155 & 15.15 & 1.91 & 5834 & -2.20 & 0.78 & 17.85 & 0.77 & 1.67 & 0.14 & 18.80 & 3.69 & 0.24 & 0.22 + 3383 & & 18.50 & 1.86 & 4640 & -2.41 & -6.56 & 21.31 & 1.80 & 1.61 & 0.11 & & & & 1.00 + 3387 & & 18.22 & 1.70 & 7433 & -2.45 & 7.79 & 23.23 & 4.97 & 2.30 & 0.14 & & & & 1.00 + 3390 & d176 & 15.89 & 1.75 & 6892 & -2.49 & 4.56 & 17.83 & 0.79 & 1.09 & 0.31 & 18.21 & 2.74 & 0.54 & 0.26 + 3392 & & 18.07 & 1.44 & & -2.62 & -0.97 & 23.09 & 6.78 & 1.49 & 0.26 & & & & 1.00 + 3400 & ic3973 d103 & 15.32 & 1.88 & 4722 & -2.65 & -5.19 & 17.80 & 1.02 & 1.92 & 0.10 & 19.34 & 3.79 & 0.41 & -0.54 + 3403 & d087 & 16.87 & 1.79 & 7790 & -2.70 & -10.75 & 22.10 & 4.95 & 5.68 & 0.07 & & & & 1.00 + 3406 & & 18.76 & 1.76 & 7166 & -2.76 & 3.01 & 21.75 & 1.93 & 1.57 & 0.17 & & & & 1.00 + 3414 & n4871 d131 & 14.89 & 1.90 & 6788 & -2.82 & -0.86 & 18.60 & 1.26 & 2.59 & 0.10 & 19.43 & 5.75 & 0.40 & 0.32 + 3423 & ic3976 d088 & 15.80 & 1.95 & 6819 & -2.97 & -7.23 & 19.40 & 1.93 & 4.27 & 0.09 & 19.25 & 3.73 & 0.40 & 0.57 + 3433 & d177 & 16.56 & 1.79 & 5542 & -3.09 & 4.19 & 18.68 & 0.84 & 1.40 & 0.31 & 19.07 & 2.52 & 0.43 & 0.30 + 3439 & d178 & 16.72 & 1.81 & 5641 & -3.15 & 6.89 & 20.81 & 1.37 & 4.81 & 0.10 & 19.97 & 2.65 & 0.14 & 0.33 + 3448 & & 18.39 & 2.28 & 29731 & -3.27 & 15.57 & 20.16 & 1.02 & 5.75 & 0.20 & & & & -1.00 + 3463 & & 18.09 & 1.87 & 6552 & -3.44 & -11.13 & 21.80 & 0.95 & 2.01 & 0.56 & 19.79 & 2.92 & 0.67 & 0.05 + 3471 & d156 & 16.45 & 9.99 & 6708 & -3.57 & 1.67 & 21.40 & 4.92 & 3.40 & 0.22 & & & & 1.00 + 3475 & & 17.21 & 1.70 & 7977 & -3.63 & 19.01 & 17.89 & 0.38 & 2.12 & 0.10 & 18.96 & 2.14 & 0.38 & 0.25 + 3481 & & 17.69 & 1.68 & 7718 & -3.72 & 12.98 & 18.92 & 0.63 & 1.69 & 0.10 & 20.55 & 2.33 & 0.06 & 0.42 + 3482 & & 18.66 & 1.85 & & -3.74 & 6.65 & 26.09 & 39.45 & 3.94 & 0.06 & & & & 1.00 + 3484 & d157 & 16.26 & 1.81 & 6119 & -3.80 & 0.17 & 21.82 & 6.45 & 5.12 & 0.10 & 21.40 & 6.66 & 0.47 & 0.81 + 3486 & & 17.73 & 1.82 & 7840 & -3.86 & -2.17 & 19.71 & 0.87 & 1.36 & 0.17 & 21.23 & 4.52 & 0.08 & 0.23 + 3487 & d132 & 16.63 & 1.88 & 7675 & -3.85 & -0.16 & 19.62 & 1.47 & 2.47 & 0.10 & 20.68 & 7.08 & 0.45 & 0.35 + 3489 & & 17.93 & 9.99 & 5506 & -3.91 & 1.57 & 24.57 & 15.69 & 4.22 & 0.20 & & & & 1.00 + 3493 & d067 & 16.50 & 1.94 & 6033 & -3.97 & -13.91 & 18.75 & 1.30 & 1.61 & 0.28 & 19.77 & 3.54 & 0.54 & 0.56 + 3565 & & 18.44 & 1.84 & 7171 & -5.09 & 0.18 & 22.85 & 5.13 & 2.33 & 0.39 & & & & 1.00 + 3567 & & 18.71 & 1.77 & & -5.10 & -0.15 & 22.41 & 3.48 & 1.74 & 0.37 & & & & 1.00 + 3615 & & 18.82 & 2.22 & 6317 & -5.86 & 0.95 & 20.78 & 0.70 & 3.64 & 0.26 & 20.21 & 1.70 & 0.25 & -0.26 + 3639 & n4867 d133 & 15.44 & 1.83 & 4805 & -6.12 & 0.02 & 21.52 & 9.11 & 6.13 & 0.28 & & & & -1.00 + 3645 & & 18.64 & 1.86 & 6417 & -6.23 & -4.50 & 20.50 & 1.25 & 1.64 & 0.08 & & & & -1.00 + 3656 & d180 & 15.53 & 1.77 & 7815 & -6.33 & 6.36 & 19.31 & 1.09 & 2.52 & 0.11 & 20.26 & 5.60 & 0.07 & 0.20 + 3660 & ic3963 d068 & 15.76 & 1.87 & 6853 & -6.49 & -11.76 & 19.45 & 1.55 & 2.65 & 0.11 & 19.96 & 5.64 & 0.35 & -0.33 + 3664 & n4864 d159 & 14.70 & 9.99 & 6812 & -6.61 & 0.38 & 19.93 & 4.97 & 2.36 & 0.10 & & & & -1.00 + 3681 & & 18.01 & 1.73 & 6913 & -6.89 & 2.32 & 21.25 & 2.25 & 1.55 & 0.05 & & & & 1.00 + 3706 & & 17.61 & 1.85 & 6891 & -7.33 & -6.19 & 19.53 & 1.16 & 3.38 & 0.13 & & & & -1.00 + 3707 & & 17.76 & 1.82 & 7212 & -7.34 & 4.22 & 21.80 & 5.03 & 2.92 & 0.58 & & & & 1.00 + 3719 & & 18.29 & 1.91 & 7812 & -7.45 & -4.38 & 21.27 & 1.11 & 2.83 & 0.17 & 21.11 & 2.51 & 0.17 & 0.34 + 3730 & ic3959 d069 & 15.27 & 1.94 & 7044 & -7.66 & -11.18 & 19.48 & 3.24 & 2.29 & 0.13 & & & & -1.00 + 3733 & ic3960 d109 & 15.85 & 1.89 & 6600 & -7.70 & -6.93 & 17.65 & 0.75 & 1.16 & 0.18 & 18.89 & 2.97 & 0.13 & 0.28 + 3739 & ic3957 d070 & 15.88 & 1.87 & 6352 & -7.85 & -12.11 & 22.55 & 9.58 & 9.59 & 0.09 & & & & -1.00 + 3750 & & 18.08 & 1.91 & 6267 & -7.99 & -11.88 & 20.04 & 1.27 & 1.24 & 0.39 & 21.82 & 4.72 & 0.00 & 0.32 + 3761 & ic3955 d160 & 15.57 & 1.88 & 7664 & -8.11 & 1.57 & 18.43 & 0.99 & 1.73 & 0.05 & 19.43 & 4.94 & 0.29 & -0.24 + 3780 & & 17.89 & 1.85 & 5080 & -8.37 & 4.79 & 22.40 & 5.40 & 2.62 & 0.49 & & & & 1.00 + 3782 & d108 & 16.55 & 1.85 & 6426 & -8.43 & -3.57 & 19.45 & 1.22 & 3.31 & 0.16 & 19.50 & 3.53 & 0.32 & 0.34 + 3794 & d134 & 17.37 & 1.98 & 6956 & -8.53 & -0.68 & 18.82 & 0.82 & 2.05 & 0.26 & 19.21 & 1.69 & 0.13 & 0.43 + 3851 & d135 & 16.98 & 1.86 & 8283 & -9.43 & -0.18 & 22.18 & 6.06 & 4.92 & 0.23 & & & & 1.00 + 3855 & & 18.05 & 1.79 & 5564 & -9.57 & -2.16 & 22.54 & 4.40 & 2.25 & 0.20 & & & & 1.00 + 3882 & d071 & 16.97 & 1.85 & 6893 & -10.00 & -11.10 & 20.70 & 2.71 & 1.72 & 0.56 & 20.34 & 5.04 & 0.66 & 0.39 + 3895 & & 17.74 & 1.72 & 8651 & -10.21 & -8.91 & 21.37 & 3.43 & 1.01 & 0.26 & & & & 1.00 + 3914 & d136 & 16.57 & 1.81 & 5661 & -10.49 & -0.35 & 19.82 & 2.33 & 3.54 & 0.14 & & & & 1.00 + 3925 & & 18.00 & 2.00 & 6448 & -10.72 & -10.48 & 21.39 & 3.28 & 1.66 & 0.04 & & & & 1.00 + 3943 & & 16.93 & 1.88 & 5507 & -11.01 & -9.42 & 19.86 & 1.28 & 1.33 & 0.29 & 19.69 & 3.63 & 0.37 & 0.21 + 3958 & ic3947 d072 & 15.94 & 1.91 & 5655 & -11.22 & -11.12 & 19.64 & 2.31 & 3.24 & 0.35 & 19.08 & 2.86 & 0.37 & 0.57 + 3972 & d181 & 16.52 & 1.87 & 6046 & -11.46 & 6.81 & 19.40 & 1.70 & 2.32 & 0.14 & 20.51 & 5.80 & 0.84 & 0.78 + 3973 & & 18.66 & 1.86 & 6715 & -11.49 & 1.50 & 23.06 & 3.97 & 3.21 & 0.18 & & & & 1.00 + 3982 & & 19.40 & 2.46 & 45675 & -11.62 & 5.01 & 20.56 & 1.15 & 3.39 & 0.18 & & & & 1.00 + 3993 & & 19.02 & 2.36 & 47234 & -11.80 & -3.58 & 22.18 & 2.49 & 5.72 & 0.28 & & & & 1.00 + 3997 & ic3946 d091 & 15.28 & 1.95 & 5879 & -11.96 & -9.60 & 17.81 & 1.30 & 1.32 & 0.36 & 19.02 & 4.96 & 0.30 & 0.29 + 4003 & & 18.60 & 1.81 & 7074 & -12.05 & 2.89 & 20.32 & 0.49 & 0.52 & 0.08 & 21.13 & 3.18 & 0.40 & 0.10 + 4024 & & 17.97 & 2.39 & 47492 & -12.43 & -4.12 & 22.13 & 1.03 & 8.65 & 0.19 & 19.62 & 2.32 & 0.22 & 0.09 + 4035 & & 18.49 & 1.82 & 6657 & -12.67 & -12.99 & 22.94 & 4.38 & 1.97 & 0.06 & & & & 1.00 + 4060 & & 17.57 & 1.31 & 8703 & -13.32 & -12.59 & 22.82 & 5.69 & 1.10 & 0.20 & & & & 1.00 + 4063 & & 19.32 & 2.38 & 47692 & -13.45 & -2.89 & 21.19 & 1.38 & 4.07 & 0.10 & & & & 1.00 + 4081 & & 18.91 & 2.15 & 50090 & -13.71 & -3.00 & 24.95 & 9.46 & 12.19 & 0.70 & 20.48 & 1.21 & 0.32 & 0.73 + 4083 & & 17.82 & 1.91 & 6178 & -13.70 & -8.60 & 21.12 & 2.49 & 1.65 & 0.55 & 21.14 & 4.34 & 0.54 & 0.44 + 4103 & & 17.74 & 1.76 & 6017 & -14.10 & -0.99 & 21.11 & 0.64 & 15.95 & 0.43 & 19.58 & 2.69 & 0.56 & -0.11 + 4118 & & 18.31 & 1.56 & 5387 & -14.31 & -8.88 & 22.82 & 5.22 & 1.10 & 0.42 & & & & 1.00 + 4129 & & 18.54 & 1.92 & 6088 & -14.60 & -7.26 & 21.60 & 2.29 & 1.95 & 0.03 & & & & 1.00 + 4141 & & 18.71 & 1.73 & & -14.74 & -5.85 & 23.46 & 7.50 & 1.24 & 0.40 & & & & -1.00 + 4175 & & 18.65 & 1.85 & 4628 & -15.26 & -8.02 & 20.16 & 0.87 & 0.92 & 0.33 & 21.12 & 3.21 & 0.45 & 0.28 + 4200 & d182 & 16.84 & 1.72 & 5638 & -15.63 & 4.75 & 20.19 & 2.36 & 2.75 & 0.01 & & & & 1.00 + 4205 & & 19.45 & 2.60 & 62822 & -15.72 & -7.05 & 20.92 & 1.57 & 3.59 & 0.41 & & & & 1.00 + lrrc type & number & @xmath109 & @xmath110 + + all & 162 & 6900 & 1182@xmath111 + ( b / [email protected] ) & 61 & 6994 & 1137@xmath113 + ( b / [email protected] ) & 68 & 6904 & 1183@xmath115 + e ( b / t=1 ) & 54 & 6864 & 1234@xmath116 + e ( @xmath117 ) & 15 & 6871 & 1129@xmath118 + e ( @xmath119 ) & 39 & 6862 & 1273@xmath120 + +
e @xmath121 & 29 & 6825 & 1035@xmath122 + e @xmath123 & 25 & 6910 & 1430@xmath124 + s0 ( 0.5@xmath125b / t@xmath1251 ) & 14 & 7057 & 942@xmath126 + | we present a quantitative morphological analysis of 187 galaxies in a region covering the central 0.28 square degrees of the coma cluster . structural parameters from the best - fitting srsic @xmath0 bulge plus , where appropriate , exponential disc model , are tabulated here .
this sample is complete down to a magnitude of @xmath1=17 mag . by examining the edwards et al .
( 2002 ) compilation of galaxy redshifts in the direction of coma , we find that 163 of the 187 galaxies are coma cluster members , and the rest are foreground and background objects .
for the coma cluster members , we have studied differences in the structural and kinematic properties between early- and late - type galaxies , and between the dwarf and giant galaxies .
analysis of the elliptical galaxies reveals correlations among the structural parameters similar to those previously found in the virgo and fornax clusters . comparing the structural properties of the coma cluster disc galaxies with disc galaxies in the field , we find evidence for an environmental dependence : the scale lengths of the disc galaxies in coma are 30% smaller .
a kinematical analysis shows marginal differences between the velocity distributions of ellipticals with srsic index @xmath2 ( dwarfs ) and those with @xmath3 ( giants ) ; the dwarf galaxies having a greater ( cluster ) velocity dispersion .
finally , our analysis of all 421 background galaxies in the catalog of edwards et al . reveals a non - uniform distribution in redshift with contrasts in density @xmath4 , characterized by a void extending from @xmath5 to @xmath6 km s@xmath7 , and two dense and extended structures centred at @xmath6 and @xmath8 km s@xmath7 . |
the mixing between different quarks is described by an unitary matrix in the charged current interaction of w - boson in the mass eigen - state of quarks , the cabibbo @xcite - kobayashi - maskawa @xcite(ckm ) matrix @xmath3 , defined by @xmath4 where @xmath5 , @xmath6 . for n - generations
, @xmath7 is an @xmath8 unitary matrix . with three generations , one can write @xmath9 a commonly used parametrization for mixing matrix with three generations of quark is given by @xcite , @xmath10 where @xmath11 and @xmath12 with @xmath13 being angles rotating in flavor space and @xmath14 is the cp violating phase .
we refer this as the pdg parametrization .
there are a lot of experimental data on the mixing pattern of quarks .
fitting available data , the mixing angles and cp violating phase are determined to be @xcite @xmath15 from the above , we obtain the magnitude of the matrix elements as @xmath16 the angles can be viewed as rotations in flavor spaces . but both the angles and the phase in the ckm matrix are not directly measurable quantities .
there are different ways to parameterize the mixing matrix . in different parametrizations ,
the angles and phase are different .
to illustrate this point let us study the original km parametrization @xcite , @xmath17 using the observed values for the mixing matrix , one would obtain @xmath18 and the central value of the cp violating phase angle is @xmath19 .
we see that the angles and phases in the pdg and km parameterizations are indeed very different .
the angles and phase are parametrization dependent .
it is interesting to see whether all quantities used to parameterize the mixing matrix can all have well defined physical meanings , that is , all are experimentally measurable quantities , as have been done for several other quantities related to mixing matrices @xcite . to this end
we notice that the magnitudes of the ckm matrix elements are already experimentally measurable quantities , one can take them to parameterize the mixing matrix
. however , the information on cp violation is then hid in the matrix elements , a single magnitude of an element would not be able to be taken as the measure of cp violation .
only a combination of several magnitude is able to signify the cp violation .
for example , @xmath20 @xcite is non - zero .
experimentally there are several measurable phases which can signify cp violations .
the famous ones are the angles @xmath0 , @xmath1 and @xmath2 in the unitarity triangle defined by the unitarity condition @xmath21 in the complex plane , the above defines a triangle shown in fig 1 .
this triangle defines three phase angles @xmath22 cp violation dictates that the area of this triangle to be non - zero .
this implies that none of the angles @xmath0 , @xmath1 and @xmath2 can be zero . experimentally these three angles have been measured directly @xcite , @xmath23 , @xmath24 and @xmath25 .
these numbers are consistent with that obtained using the numerical numbers in eq .
[ qangle ] , @xmath26 , @xmath27 and @xmath28 .
also the directly measured numbers are consistent with the sm prediction @xmath29 .
notice that the values @xmath0 , @xmath2 are very close to the two phases @xmath30 , @xmath14 , respectively .
we will see later that although they are close to each other , they are not exactly equal .
since we know that one only needs one quantity to signify the existence of cp violation , @xmath0 , @xmath1 and @xmath2 must be related .
in fact they are related to the parameter @xmath31 as @xcite @xmath32 the above is twice of the triangle area in fig 1 .
the quantity @xmath33 is equal to @xmath34 . to have a parametrization for @xmath35 in which all quantities are experimentally measurable ones
, one can choose three modulus of @xmath36 and one of the above cp violating parameters .
for the cp violating phase , it is clear that the three phase angles @xmath0 , @xmath1 and @xmath2 are among the simplest and have clear geometric meaning .
we refer these as the @xmath0 , @xmath1 and @xmath2 parameterizations . in the following
we discuss how the parametrization can be constructed and how they can be transformed from each other .
in the @xmath0 parametrization we take @xmath0 as the phase appearing in the ckm matrix along with three modulus of @xmath37 . from the definition of @xmath0 in eq.[phases ] , we can have four ways in which only one of the @xmath38 relevant to the definition of @xmath0 is complex and all others are real and positive , @xmath39 in the above one can change the signs of the elements by defining quark phases . indicating the ckm matrix for the four cases by @xmath40 .
we have @xmath41 one can take @xmath0 , @xmath42 , @xmath43 , @xmath44 as the four independent variables to parameterize the ckm matrix .
the other elements can be expressed as functions of them with @xmath45 for the other three cases , @xmath35 are given by @xmath46 similar to case @xmath47 , one can choose the phase @xmath0 and three modulus of @xmath37 as independent variables for the above three cases .
it is convenient to choose the parameter sets ( @xmath0 , @xmath43 , @xmath48 , @xmath49 ) , ( @xmath0 , @xmath50 , @xmath44 , @xmath51 ) and ( @xmath0 , @xmath49 , @xmath52 , @xmath50 ) as independent variables for the above three cases , respectively . in all the above four cases ,
the jarlskog parameter @xmath31 is given by @xmath53 this is not surprising because the above four cases are equivalent . to see the above four cases discussed
are equivalent explicitly , let us demonstrate how one can transform case @xmath47 to case @xmath54 by redefining quark phases .
the different parameterizations are equivalent implies that by redefining quark phases , one can transform the different ways of parameterizations for @xmath55 from one to another , that is , @xmath56 where @xmath57 and @xmath58 stand for various types of parametrization .
for example , transforming case @xmath47 to case @xmath54 becomes a mission of finding the different parameters such that + @xmath59 comparing the coefficients , one obtains @xmath60 , @xmath61 , and @xmath62}\over{2|v_{ud}||v_{ub}||v_{cd}||v_{cb}|}}\;.\end{aligned}\ ] ] therefore , the transformation from case @xmath47 to case @xmath54 is achieved by , @xmath63 similarly one can transform the other two cases to case @xmath54 too .
we now discuss the @xmath1 parametrization . from eq.[phases ] , we can have four choices for the location of the phase similar to the @xmath0 parametrization .
they are @xmath64 in a similar way as for the four cases of @xmath65 , one can show that the above four cases are equivalent .
detailed expressions for these four cases are given in the appendix .
we will display the explicit form for case @xmath66 .
we have @xmath67 for this case it is convenient to take @xmath1 , @xmath68 , @xmath49 , @xmath52 as independent variables .
the other quantities can be expressed as @xmath69 the jarlskog parameter @xmath31 is given by @xmath70 for @xmath2 parametrization , the four different cases are defined by @xmath71 the detailed expressions for these four cases are given in appendix .
we display the explicit form for case @xmath72 here .
we have @xmath73 taking @xmath2 , @xmath44 , @xmath68 , @xmath51 as variables , the other quantities can be expressed as @xmath74 and @xmath75
we now show that the @xmath0 , @xmath1 and @xmath2 parameterizations can also be transformed from each other .
the twelve parameterizations discussed before in the previous section are all equivalent .
we have already shown that the four parameterizations for @xmath0 or @xmath1 or @xmath2 are equivalent ones , we will therefore only need to show that one of the @xmath0 parameterizations is equivalent to one of the @xmath1 or @xmath2 parameterizations .
we now show that @xmath76 and @xmath77 .
for these two parametrizations , the first column and the third row are already identical in these two parameterizations . using @xmath78
one can readily show that the 12 , 13 , 22 and 23 entries of @xmath79 and @xmath77 are equal .
therefore @xmath80 similarly , using @xmath81 we therefore have shown explicitly that the @xmath0 , @xmath1 and @xmath2 parameterizations are related and can be transformed from one to another .
numerically , one finds that the approximate relations @xmath82 and @xmath83 .
these can be understood easily by noticing the relations between them @xcite , @xmath84 therefore , @xmath85 is approximately @xmath86 , since @xmath0 is close to @xmath87 , @xmath30 must also be close to @xmath87 and therefore @xmath82 .
it is also clear that @xmath14 is approximately equal to @xmath2 .
one may wonder if there is a parametrization where the phase is close to @xmath1 .
we find indeed there are angle prametrizations in which the cp violating phase is close to @xmath1 .
an example is provided by the parametrization @xmath88 discussed in ref .
@xcite where @xmath89 we have @xmath90 one may also wonder if one can find a parametrization in which the cp violating angle is one of the @xmath0 , @xmath1 and @xmath2 , and the other three quantities to parameterize the mixing matrix can be chosen to be three angles similar to those in the pdg or km paramerizations .
we find that this is impossible .
it has been shown that there are nine different ways to parameterize the mixing matrix using one phase and three angles @xcite .
explicit inspections find that none of the phases can be exactly , allowing sign differences or plus or minus a @xmath86 , identified as one of the @xmath0 , @xmath1 or @xmath2 .
the reason is that in the @xmath0 , @xmath1 and @xmath2 parametrization we have introduced , one needs one of the elements in mixing matrix to be of the form @xmath91 , @xmath92 , or @xmath93 , and also another five real matrix elements which can not be satisfied with only three angles .
it has proven to be convenient to use approximate formula such as the wolfenstein parametrization@xcite . in the literature different approximate forms
have been proposed@xcite . in this section ,
we discuss the wolfenstein - like expansions in the @xmath0 , @xmath1 and @xmath2 parameterizations demanding to use one of the @xmath0 , @xmath1 and @xmath2 as one of the parameters . for the @xmath94 cases ,
we find it convenient to work with @xmath95 case .
we use @xmath96 , @xmath97 , @xmath98 and @xmath0 as parameters .
the numerical values of @xmath99 , @xmath100 and @xmath101 are determined to be @xmath102 , @xmath103 , and @xmath104 .
to order @xmath105 , we have @xmath106 one can further rotate the phase of c - quark by @xmath86 and b - quark by @xmath107 to obtain an expression where the diagonal entries are close to 1 .
we obtain @xmath108 the above expansion is equivalent to that discussed in ref.@xcite .
the parameters @xmath109 , @xmath110 and @xmath111 in ref.@xcite are related to the above parameters by , @xmath112 , @xmath113 and @xmath114 . at more than @xmath105 order , there are differences for our approximation and that proposed in ref .
@xcite . for @xmath115 cases
, it is convenient to use @xmath116 for expansion . setting @xmath117 , @xmath118 , and @xmath119 with @xmath120 , and @xmath121 @xmath122 . rotating the b - quark field by a phase @xmath123
, we obtain to order @xmath105
@xmath124 for the @xmath125 cases ,
@xmath126 case is convenient for expansion . setting @xmath117 , @xmath127 and @xmath128 with @xmath129 , @xmath130 , and @xmath122 .
rotating d - quark by a phase @xmath86 and u - quark by a phase @xmath131 , we obtain to order @xmath105 @xmath132 to @xmath105 order , the above expansion is equivalent to the wolfenstein parametrization@xcite .
the parameters @xmath133 , @xmath134 and @xmath135 in the wolfenstein parametrization are related to @xmath100 , @xmath136 and @xmath2 by @xmath137 , @xmath138 and @xmath139 .
again at more than @xmath105 order , there are differences in these two approximations . among the three phase angles , @xmath0 , @xmath1 and @xmath2 , the best measured one is @xmath1 .
this makes the approximate expression @xmath140 better than others .
to conclude , we have proposed new parametrizations of the ckm matrix using the three measurable phase angles @xmath0 , @xmath1 and @xmath2 in the unitarity triangle as the cp violating phase . for each of the @xmath0 , @xmath1 and @xmath2 parametrization
there are four ways to parameterize the mixing matrix where one column and one row elements are all real .
we have shown explicitly that all these cases are equivalent .
we have studied relations of the @xmath0 , @xmath1 and @xmath2 parametrizations with the usual three rotation angles in flavor space and one cp violating phase parametrizations .
we find that it is not possible to parameterize the ckm matrix using three angles and taking one of the @xmath0 , @xmath1 and @xmath2 as the cp violating phase
. however , there are rotation angle parametrizations in which the cp violating are very close to one of the phase angles @xmath0 , @xmath1 or @xmath2 . we , however , emphasis that the @xmath0 , @xmath1 and @xmath2 parametrizations we proposed have the advantage that all the elements used to describe the mixing matrix are physically measurable quantities , unlike the parametrizations of using rotation angles in flavor space and a phase whose values are parametrization dependent .
the @xmath0 , @xmath1 and @xmath2 parametrizations are parametrization independent representation of the ckm matrix .
we also suggest new wolfenstein - like paramterizations .
+ * acknowledgement * this work was supported in part by nsc and ncts of roc , nnsf(grant no:11175115 ) and shanghai science and technology commission ( grant no : 11dz2260700 ) of prc .
expressions for the four @xmath1 and four @xmath2 parameterizations . p. frampton and xiao - gang he , phys .
lett . * b688*:67 - 70(2010)[arxiv:1003.0310 [ hep - ph ] ] ; phys . rev . *
d82*:017301(2010)[arxiv:1004.3679 [ hep - ph ] ] ; s. -w .
li and b. -q .
ma , phys .
b * 691 * , 37 ( 2010 ) [ arxiv:1003.5854 [ hep - ph ] ] .
n. li and b. -q .
ma , phys .
d * 71 * , 097301 ( 2005 ) [ hep - ph/0501226 ] ; x. -g .
he , s. -w .
li and b. -q .
ma , phys .
d * 78 * , 111301 ( 2008 ) [ arxiv:0809.1223 [ hep - ph ] ] ; phys .
d * 79 * , 073001 ( 2009 ) [ arxiv:0903.2880 [ hep - ph ] ] ; y. h. ahn , h. -y .
cheng and s. oh , phys . lett .
b * 703 * , 571 ( 2011 ) [ arxiv:1106.0935 [ hep - ph ] ] .
n. qin and b. -q .
ma , phys .
b * 695 * , 194 ( 2011 ) [ arxiv:1011.6412 [ hep - ph ] ] ; n. qin and b. -q .
ma , phys .
d * 83 * , 033006 ( 2011 ) [ arxiv:1101.4729 [ hep - ph ] ] ; y. h. ahn , h. -y .
cheng and s. oh , phys .
b * 701 * , 614 ( 2011 ) [ arxiv:1105.0450 [ hep - ph ] ] . | the ckm matrix describing quark mixing with three generations can be parameterized by three mixing angles and one cp violating phase . in most of the parameterizations ,
the cp violating phase chosen is not a directly measurable quantity and is parametrization dependent . in this work ,
we propose to use experimentally measurable cp violating quantities , @xmath0 , @xmath1 or @xmath2 in the unitarity triangle as the phase in the ckm matrix , and construct explicit @xmath0 , @xmath1 and @xmath2 parameterizations .
approximate wolfenstein - like expressions are also suggested . |
quantum information has changed not only the way we understand quantum mechanics ; it has also changed the way we use quantum mechanics in our dealings with the real world . for years
, the research on quantum mechanics has focused on pointing out how different quantum mechanics was from classical physics .
the next step has been to realize that quantum mechanics can be used to solve problems unsolvable by any other means . however , although quantum states are powerful tools , they are very fragile for most real - life applications . therefore , the challenge is twofold : on one hand , to find out new applications
problems without classical solutions that can be solved with the aid of quantum states ; on the other , to find out how to protect quantum states methods that allow us to use quantum states for practical purposes . in this paper
i review some results regarding a particular family of quantum states .
the properties of these states , particularly their symmetry , grant them a prominent role in this challenge .
supersinglets @xmath0 are states of total spin zero of @xmath1 particles , all of them of spin @xmath5 .
supersinglets are @xmath1-lateral rotationally invariant .
this means that , if we act on any of them with the tensor product of @xmath1 equal rotation operators , the result will be to reproduce the same state ( within a possible phase factor ) : @xmath6 @xmath7 being @xmath8 , where @xmath9 is a rotation operator . in this paper
, @xmath10 denotes the tensor product state @xmath11 , where @xmath12 corresponds to the state with eigenvalue @xmath13 of the spin along the @xmath14-direction .
the basis we shall use , @xmath15 , and the basis of eigenvectors of the spin along the @xmath14-direction , @xmath16 , have the same kets , except that they are in reverse order : the ket @xmath17 of @xmath18 is equal to the ket @xmath19 of @xmath20 .
the interest in some types of supersinglets is not new .
for instance , the two - particle @xmath2-level supersinglets @xmath21 attracted some attention in connection with violations of bell s inequalities for systems of two spin-@xmath22 particles @xcite .
these states can be expressed as @xmath23 examples of @xmath24 states are the following : @xmath25 the @xmath26 state has also been used in two - particle proofs of bell s theorem without inequalities @xcite .
a method for preparing optical analogs of @xmath27 for every @xmath2 has been recently described @xcite .
also , an experimental violation of a bell s inequality using an optical analog of @xmath28 has been recently reported @xcite . in this paper
we are interested in two types of supersinglets : @xmath1-particle @xmath1-level supersinglets , @xmath3 , and @xmath1-qubit ( with @xmath1 even ) supersinglets , @xmath29 .
the former can be expressed as @xmath30 where @xmath31 is the number of transpositions of pairs of elements that must be composed to place the elements in canonical order ( i.e. , @xmath32 ) .
the @xmath33 state is given in equation ( [ s2s2 ] ) .
other examples of @xmath34 states are the following : @xmath35 the @xmath36 state has been used in the solution of the so - called _ byzantine agreement problem _
@xcite ( see more on this problem in section [ sec:3 ] ) .
the @xmath37 states have been used in a scheme designed to probe a quantum gate that realizes an unknown unitary transformation @xcite .
some interesting properties of @xmath38 have been reported in @xcite .
an experimental realization of @xmath36 has been recently proposed @xcite . for @xmath1
even , the @xmath1-qubit supersinglet can be expressed as @xmath39 where the sum is extended to all the states obtained by permuting the state @xmath40 , which contains the same number of zeros and ones ; @xmath14 is the number of zeros in the first @xmath41 positions ( for example : in @xmath42 , @xmath43 ; in @xmath44 , @xmath45 ; in @xmath46 , @xmath47 ) .
the @xmath48 state is given in equation ( [ s2s2 ] ) .
other examples of @xmath4 ( with @xmath1 even ) states are the following : @xmath49 very recently , the @xmath50 state has been experimentally prepared using four parametric down - converted photons entangled in polarization @xcite .
some applications of the @xmath51 states are described in section [ sec:5 ] .
the structure of this paper is the following : in section [ sec:2 ] , three apparently unrelated problems are introduced . in section [ sec:3 ] , these three problems are proved to not have classical solution . in section [ sec:4 ] a solution based on the use of the @xmath37 states is presented .
finally , in section [ sec:5 ] , the use of the @xmath51 states to create decoherence - free subspaces with qubits is described .
the scenario for the _ @xmath1 strangers problem _ ( nsp ) is an extension to a high number @xmath1 of players of the situation described in patricia highsmith s novel @xcite and alfred hitchcock s movie @xcite _ strangers on a train _ : @xmath1 complete strangers @xmath52 ( @xmath53 ) meet on a train .
@xmath52 wants @xmath54 dead .
in the course of small talk , one suggests that an `` exchange '' murder between @xmath1 complete strangers would be unsolvable .
after all , how could the police find the murderer when he / she is a total and complete stranger with absolutely no connection whatsoever to the victim ?
@xmath52 could kill @xmath55 , etc .
however , such a plan suffers from an important problem : if all the players know who the murderer of each victim is , then the whole plan is vulnerable to individual denunciations . alternatively ,
if the distribution of victims is the result of a secret lottery , how could the murderers be assured that the lottery was not rigged and that nobody had contrived the result or could ascertain it ? in more general , noncriminal , terms
the @xmath1 strangers problem can be reformulated as follows : each of @xmath1 parties must perform one among @xmath1 different tasks which they do not want to perform themselves . however , each party volunteers to perform some other party s task .
the problem is then how to distribute the victims / tasks @xmath56 among the murderers / volunteers @xmath57 , which share no previous secret information nor any secure classical channel , in a way that guarantees that each murderer / volunteer @xmath52 knows only the identity of his / her victim / task and that nobody else ( besides the murderers / volunteers ) knows anything about the victim / task assignments .
the _ secret sharing problem _ ( ssp ) was already described , for @xmath58 , in @xcite .
it could arise in the following context : @xmath59 wants to have a secret action taken on her behalf at a distant location .
there she has @xmath60 agents , @xmath61 , @xmath62 , , @xmath63 who carry it out for her .
@xmath59 knows that some of them are dishonest , but she does not know which ones .
she can not simply send a secure message to all of them , because the dishonest ones will try to sabotage the action , but it is assumed ( as in @xcite ) that if all of them carry it out together , the honest ones will keep the dishonest ones from doing any damage .
the problem is then that @xmath59 wishes to convey a cryptographic key to @xmath61 , @xmath62 ,
, @xmath63 in such a way that none of them can read it on their own , only if all the @xmath52 ( @xmath64 ) collaborate .
in addition , they wish to prevent any eavesdropper from acquiring information without being detected .
it is assumed that @xmath59 shares no previous secret information nor any secure classical channel with her agents .
different quantum solutions to this problem for @xmath58 has been proposed using either ghz @xcite or bell states @xcite ( see also @xcite ) .
below we shall propose a different solution for any @xmath1 .
the _ liar detection problem _ ( ldp ) occurs in the following scenario : three parties @xmath65 , @xmath66 , and @xmath67 are connected by secure pairwise classical channels .
let us suppose that @xmath65 sends a message @xmath68 to @xmath66 and @xmath67 , and @xmath66 sends the same message to @xmath67 .
if both @xmath65 and @xmath66 are honest , then @xmath67 should receive the same @xmath68 from @xmath65 and @xmath66 .
however , @xmath65 could be dishonest and send different messages to @xmath66 and @xmath67 , @xmath69 , or , alternatively , @xmath66 could be dishonest and send a message which is different from the one he receives , @xmath70 .
for @xmath67 the problem is to ascertain without any doubt who is being dishonest .
this problem is interesting for classical information distribution in pairwise connected networks .
the message could be a database and the dishonest behavior a consequence of an error during the copying or distribution process .
in this section i shall explain in which sense each of the three problems introduced in section [ sec:2 ] have no classical solution .
i will furthermore offer proofs of the impossibility of the existence of that kind of solutions . in peres book @xcite ( p. 293 )
, the _ key distribution problem _ ( kdp ) is defined as follows : `` the problem ( ) is how to distribute a cryptographic key ( a secret sequence of bits ) to several observers who initially share no secret information , by using an insecure communication channel subject to inspection by a hostile eavesdropper .
if only classical means are used , this is an impossible task .
quantum phenomena , on the other hand , provide various solutions .
the reason for this difference is that information stored in classical form ( ) can be examined objectively without altering it in any detectable way , let alone destroying it , while it is impossible to do that with quantized information encoded in unknown non - orthogonal states . '' in this definition , trusted couriers , clandestine meetings or private secure communication links between the parties are not allowed .
it is in this scenario namely , one in which previous secret information and secure communication channels between all parties ( in the nsp ) or between @xmath59 and her agents ( in the ssp ) are not allowed , but public classical channels which can not be altered are assumed in which the first two problems ( nsp and ssp ) have no classical solution .
a proof follows .
the kdp between two parties has no classical solution . _
proof : _ information stored in classical form can be examined objectively without altering it in any detectable way , _
. _ the nsp has no classical solution . _
proof : _ suppose the nsp had a classical solution for some @xmath71 . then , @xmath72 parties could publicly announce the names of their victims and then the remaining two parties would have a way to solve the kdp ( for instance , by publicly assigning 0 and 1 to the remaining two victims and agreeing to define the corresponding entry of the key as the bit value assigned to the victim of one of the two parties ) , _ q.e.d . _
the ssp has no classical solution . _
proof : _ suppose the ssp had a classical solution for some @xmath73 ( if @xmath74 , then the ssp is the kdp ) .
such a solution must work even when all of @xmath59 s agents are honest . in this case
, @xmath72 agents could publicly collaborate and share the result of such a collaboration with the remaining agent .
then @xmath59 and the remaining agent would have a way to solve the kdp , _
q.e.d . _ + the scenario of the third problem , the ldp , is different .
@xmath65 , @xmath66 , and @xmath67 are now connected by pairwise classical channels .
these channels must be secure . in this scenario
namely , one in which previous secret information is not allowed , but secure pairwise classical channels are assumed , the ldp has no classical solution . in order to prove this , we need some previous results .
the _ byzantine generals problem _ ( bgp ) is defined @xcite as follows : a commanding general must send an order to his @xmath60 lieutenant generals such that : ( ic1 ) all loyal lieutenant generals obey the same order .
( ic2 ) if the commanding general is loyal , then every loyal lieutenant general obeys the order he sends .
( ic1 ) and ( ic2 ) are known as the interactive consistency conditions .
note that if the commanding general is loyal , ( ic1 ) follows from ( ic2 ) .
however , the commanding general may be a traitor .
the bgp for @xmath58 generals and @xmath75 traitor has no classical solution .
_ proof _ @xcite ( p. 384 ) : for simplicity , we consider the case in which the only possible orders are `` attack '' or `` retreat '' .
let us first examine the scenario in which the commanding general is loyal and sends an `` attack '' order , but lieutenant general 2 is a traitor and reports to lieutenant general 1 that he received a `` retreat '' order .
for ( ic2 ) to be satisfied , lieutenant general 1 must obey the order to attack . now
consider another scenario in which the commanding general is a traitor and sends an `` attack '' order to lieutenant general 1 and a `` retreat '' order to lieutenant general 2 .
lieutenant general 1 does not know who the traitor is , and he can not tell what message the commanding general actually sent to lieutenant general 2 . hence , the scenarios in these two cases appear exactly the same to lieutenant general 1 . if the traitor lies consistently , then there is no way for lieutenant general 1 to distinguish between these two situations , so he must obey the `` attack '' order in both of them .
hence , whenever lieutenant general 1 receives an `` attack '' order from the commanding general , he must obey it , _
q.e.d . _ + for a more rigorous proof , see @xcite .
no solution for the bgp for @xmath76 generals can cope with @xmath77 traitors . _ proof : _ the bgp for @xmath78 generals and @xmath77 traitors can be reduced to the bgp for @xmath58 generals and @xmath79 traitor , with each of the generals simulating at most @xmath77 lieutenants and taking the same decision as the loyal lieutenants they simulate , _
q.e.d . _ + a solution for the bgp for @xmath80 and up to @xmath77 traitors
is given in @xcite .
the protocol proposed in @xcite solves the bgp for @xmath58 in the following sense : ( 1 ) if all generals are loyal , then the protocol achieves broadcast .
( 2 ) if one general is a traitor , then either the protocol achieves broadcast or both loyal generals abort the protocol .
the ldp described in section [ sec:2 ] is simplification of the bgp for @xmath58 and @xmath79 .
the ldp has no classical solution . _
proof : _ supposing that the ldp had a classical solution , then the bgp for @xmath58 and @xmath79 had classical solution , _
in this section i will show that all three problems introduced in section [ sec:2 ] can be solved if each of the @xmath1 participants are in possession of a sequence of numbers with the following properties : * it is _ random _
( i.e. , generated by an intrinsically unrepeatable method which gives each possible number with the same probability of occurring ) .
* the possible numbers are integers from @xmath81 to @xmath60 . *
if a number @xmath82 is at position @xmath83 of the sequence of party @xmath84 , @xmath82 is not at position @xmath83 in the sequence of a different party . *
each party knows only his / her own sequence . *
nobody else ( besides the parties ) knows the sequences
. properties ( iv ) and ( v ) are difficult to accomplish using classical tools due to the fact that information transmitted in classical form can be examined and copied without altering it in any detectable way . however , as quantum key distribution protocols show @xcite , quantum information does not suffer from such a drawback . assuming we have a reliable method of generating sequences of numbers with properties ( i ) to ( v ) among @xmath1 distant parties , a method that will be presented below , then the solutions to the above problems
are described in the following subsections .
each victim @xmath54 is assigned a label , taken from @xmath81 to @xmath60 .
if murderer @xmath52 s sequence starts with @xmath83 , then @xmath52 must kill @xmath85 , etc .
the remaining entries of the sequence can be used for subsequent rounds .
the result tells every murderer who his / her victim is in such a way that prevents any murderer ( or even a small group of them ) from denouncing or blackmailing another .
the only way to ascertain with certainty who murdered @xmath85 is that all the other murderers confess .
@xmath52 has a probability of @xmath86 to have to kill his / her own desired victim @xmath54 . in case
@xmath52 must kill @xmath54 , since the fact that @xmath52 wants @xmath54 dead is presumably known by others , the best thing @xmath52 could do is to participate in another round with a different set of strangers .
the key is defined as @xmath59 s sequence . the only way to reveal it is to make the remaining @xmath60 parties share their respective sequences ;
the key is then composed by the missing results .
if a dishonest party @xmath87 declares a result which is different to his / her actual result , then there is a probability @xmath88 , where @xmath89 is the number of honest parties which have not yet declared their results , that other honest party @xmath90 has obtained that result
. then @xmath90 would stop the process , so alice s key ( and thus alice s action ) would remain safe ( dishonest parties can not sabotage alice s action if they do not know what it is ) .
the order in which the agents declare their respective results must change from round to round to avoid any dishonest party being always the last to declare . before entering into the details of the solution
, it will be useful to sketch how it works .
@xmath65(lice ) must send @xmath66(ob ) [ and @xmath67(arol ) ] some information @xmath91 [ @xmath92 such that : ( a ) it is only known to alice .
( b ) its authenticity can be checked by @xmath66 ( and @xmath67 ) using information only known to him ( her ) .
( c ) the information @xmath91 [ @xmath92 is correlated with the message @xmath93 ( @xmath94 ) ; if the message were different , then this information would also be different .
bob needs @xmath91 to convince carol that the message @xmath65 sent him was actually @xmath93 ( and not other ) .
protocol for solving the liar detection problem . to explain how it works
, we will assume that nobody is lying .
@xmath65 must send @xmath66 a message @xmath93 and the list @xmath91 of positions in @xmath65 s list @xmath95 ( left ) in which the message @xmath93 appears .
analogously , @xmath65 must send @xmath67 a message @xmath94 and the corresponding list @xmath96 .
finally , @xmath66 must send @xmath67 a message @xmath97 and two lists : @xmath98 and @xmath91.,width=309 ] let us suppose that the message @xmath68 is a trit value @xmath81 , @xmath75 , or @xmath99 .
all three parties agree to use the following protocol : * if the transmitted message is @xmath100 , then the sender @xmath82 must also send @xmath83 the list @xmath101 of positions in his / her sequence in which the number @xmath100 appears ( see figure [ liard ] ) . note that if the sequences are random and long enough then any @xmath101 must contain about one third of the total length @xmath102 of the sequences .
* the receiver @xmath83 would not accept any message if the intersection between the received list @xmath101 and his / her list @xmath103 is not null nor if @xmath104 elements .
we will assume that requirements ( i ) and ( ii ) force the dishonest party to send correct but perhaps incomplete lists .
otherwise , if @xmath82 sends a list containing @xmath105 erroneous data , then the probability that @xmath83 does not accept the message @xmath100 would be @xmath106 .
in addition , * @xmath66 must send @xmath67 the list @xmath107 containing the sequence he has ( supposedly ) received from @xmath65 ( see figure [ liard ] ) .
therefore , when @xmath67 finds that @xmath108 , she has received three lists to help her to find out whether it is @xmath65 or @xmath66 who is being dishonest ( see figure [ liard ] ) . according to rules ( i ) to ( iii ) , if @xmath66 wants to be dishonest @xmath109 must necessarily be a subset of @xmath110 , because @xmath66 does not know @xmath111 .
however , the length of @xmath110 is about @xmath112 , while @xmath67 is expecting @xmath66 to send her two lists with a total length of @xmath113 ; then @xmath67 would conclude that @xmath66 was being dishonest .
alternatively , if it is @xmath65 who is being dishonest , the length of the two lists that @xmath67 received from @xmath66 would total about @xmath113 ; @xmath67 would then conclude that @xmath65 was being dishonest . a possible quantum method to generate sequences of numbers with properties ( i ) to ( iv ) among @xmath1 distant parties
is by distributing among all @xmath1 parties an @xmath1-particle @xmath1-level supersinglet @xmath3 described in ( [ snsn ] ) .
once the particles have been distributed among the parties , a direction of measurement is randomly chosen and publicly announced . due to the @xmath1-lateral rotational invariance of the @xmath3 states ,
whenever the @xmath1 parties measure the spin of the @xmath1 separated particles along this direction , each of them finds a different result in the set @xmath114 ; thus such results satisfy requirements ( ii ) and ( iii ) . in order to accomplish requirements ( i ) ,
( iv ) , and ( v ) , an essential property is _ nonseparability _ , that is , the quantum predictions for the @xmath3 states can not be reproduced by any local hidden variables model in which the results of the spin measurements are somehow determined before the measurement . to show the nonseparability of @xmath3 we have to study whether they violate bell s inequalities derived from the assumptions of local realism .
most bell s inequalities require two alternative local dichotomic ( taking values @xmath115 or @xmath75 ) observables @xmath116 and @xmath85 on each particle @xmath83 . to test nonseparability
, we will use the dichotomic local observables proposed by peres in @xcite .
a peres observable @xmath117 can be operationally defined as follows : to measure @xmath117 , first measure the spin component of particle @xmath84 along direction @xmath65 , @xmath118 .
if particle @xmath84 is a spin-@xmath22 particle , then measuring @xmath118 could give @xmath119 different results .
then assign value @xmath75 to results @xmath22 , @xmath120 , etc . , and value @xmath115 to results @xmath121 , @xmath122 , etc .
the operator representing observable @xmath117 can be written as @xmath123 where @xmath124 is the eigenstate of the spin component along direction @xmath65 of particle @xmath84 . to show the nonseparability of the @xmath3 states ,
let us consider the following scenario : @xmath1 distant observers share @xmath1 @xmath1-level particles in the @xmath3 state ; the @xmath125 observers can choose between measuring @xmath126 and @xmath127 ; the remaining @xmath68 observers can choose between measuring @xmath128 and @xmath129 .
then nonseparability can be tested by means of the following bell s inequality , which generalizes to @xmath1 particles the clauser - horne - shimony - holt inequality @xcite @xmath130 note that this inequality uses only a subset of all possible correlation functions ( for instance , it does not use @xmath131 ) . for the @xmath3 states , restricting our attention to peres observables , the correlation function @xmath132 , which represents the expected value of this product of the results of measuring , for instance , @xmath125 observables @xmath65 , and @xmath68 observables @xmath66 is given by @xmath133 \over \sin \theta_{ab}}\right\},\end{aligned}\ ] ] where @xmath134 is the angle between directions @xmath65 and @xmath66 and @xmath135 gives the greatest integer less than or equal to @xmath136 . in case of @xmath137 ,
that is , using correlation functions of the @xmath138 type , we have found that @xmath3 states violate inequality ( [ chshg ] ) for any @xmath1 .
the maximum violation for @xmath139 is @xmath140 , for @xmath58 is @xmath141 , and for @xmath142 tends to @xmath143 . in case of @xmath144 , that is , using correlation functions of the @xmath145 type , we have found that the @xmath3 states violate inequality ( [ chshg ] ) for any @xmath1 .
the maximum violation for @xmath146 is @xmath147 , for @xmath148 is @xmath149 and for @xmath142 tends to @xmath143 .
so far we have assumed that the @xmath1 parties share a large collection of @xmath1-level systems in the @xmath3 state . to make sure that this is the case ( and that nobody has changed the state ) requires a protocol to distribute and test these states between the @xmath1 parties such that at the end of the protocol either all parties agree that they share a @xmath3 state ( and then they can reliably apply the described solutions ) , or all of them conclude that something went wrong ( and then abort any subsequent action ) .
for @xmath58 such a distribute - and - test protocol is explicitly described in @xcite and can be easily generalized to any @xmath150 .
the test requires that the parties compare a sufficiently large subset of their particles which are subsequently discarded .
@xmath1-qubit ( with @xmath1 even ) supersinglets @xmath4 have found many applications : they can be used to solve the kdp , and the ssp , defined in section [ sec:2 ] ; they can be used to do _ telecloning _ ( a process combining quantum teleportation and optimal quantum cloning from one input to @xmath77 outputs @xcite ) .
in fact , @xmath4 are the @xmath1-lateral rotationally invariant version of the telecloning states introduced in @xcite .
recently , it has been found that @xmath151 can also be used to solve the ldp described in section [ sec:2 ] @xcite . here , however , we shall be examining a different use for these states : encoding quantum information in a way that makes sure that it is not affected by some kind of otherwise unavoidable errors . in quantum computation and quantum communication @xcite
, information is stored in the quantum state of quantum systems , usually qubits .
preserving the desired states of the qubits and controlling their evolution is essential for most of the tasks .
however , when the qubits interact with the environment , although the overall evolution of the state of the qubits - environment system is unitary ( according to quantum mechanics , the evolution of the state of any closed system is unitary ) , the evolution of the state of the qubits alone ( which do not form a closed system ) is not unitary .
the state of the qubits rapidly evolves into a state which is entangled with the environment .
this implies that , even if the initial state of the qubits was pure ( @xmath152 ) , after such an interaction the state of the qubits rapidly becomes mixed ( @xmath153 ) .
this process is know as _ decoherence _ @xcite .
decoherence is thus a major obstacle in quantum computation and quantum communication .
there are two ways to overcome decoherence .
one is based on adding redundancy when encoding information in order to detect and correct errors by using _ quantum error correction codes _
a quantum error - correcting code is defined as a unitary mapping ( encoding ) of @xmath84 qubits into a subspace of the quantum state space of @xmath105 qubits such that , if any @xmath31 of the qubits undergos arbitrary decoherence , not necessarily independently , the resulting @xmath105 qubits can be used to faithfully reconstruct the original quantum state of the @xmath84 encoded qubits .
the other approach is by encoding qubits within subspaces which do not suffer decoherence due to reasons of symmetry .
these subspaces are called _ decoherence - free ( df ) subspaces _ @xcite . indeed , df subspaces can be considered as a special class of quantum error correction codes @xcite .
let us illustrate the use of the @xmath4 states for constructing df subspaces by considering a simple example : @xmath1 qubits coupled by a single thermal bath described by a collection of noninteracting linear oscillators .
let us assume that all qubits suffer the _ same _ interaction with the bath .
such an assumption is justified as long as the qubits have very close positions with respect to the bath coherence length .
then , it has been shown that , for @xmath1 even , the @xmath4 states are eigenstates of the whole hamiltonian of the qubits - bath system and also eigenstates of the interaction hamiltonian with eigenvalue zero ( see @xcite for details ) .
this is due to the fact that the @xmath4 states are @xmath1-lateral unitary invariant .
this means that , if we act on any of them with the tensor product of @xmath1 equal unitary operators , the result will be to reproduce the same state : @xmath154 @xmath155 being @xmath156 , where @xmath157 is an unitary operator @xcite .
in addition , any ( coherent or incoherent ) superposition of @xmath1-lateral unitary invariant states is also @xmath1-lateral unitary invariant .
therefore , for arbitrary @xmath1 ( even ) , the @xmath158 states with @xmath83 ranging from 2 to @xmath1 ( and tensor products thereof ) span a @xmath159-dimensional df subspace .
let us call this subspace @xmath160 .
the dimension of @xmath160 is given @xcite by @xmath161 the number of qubits encoded in @xmath160 is @xmath162 . for large @xmath1 , @xmath163 therefore , the encoding efficiency , defined as @xmath164 is asymptotically _
unity_. i thank n. d. mermin and c. serra for useful comments , m. bourennane , m. eibl , s. gaertner , n. kiesel , and h. weinfurter for illuminating discussions , m. ferrero and the rest of organizers for inviting me to the oviedo conference on quantum information , h. weinfurter for his hospitality at ludwig - maximilians - universitt , mnchen , and max - planck - institut fr quantenoptik , garching , and the spanish ministerio de ciencia y tecnologa grants bfm2001 - 3943 and bfm2002 - 02815 , the junta de andaluca grant fqm-239 , and the max - planck - institut fr quantenoptik for support .
m. hillery , v. buek , and a. berthiaume , phys .
a * 59 * , 1829 ( 1999 ) .
m. ukowski , a. zeilinger , m.a .
horne , and h. weinfurter , acta phys . pol .
* 93 * , 187 ( 1998 ) .
a. karlsson , m. koashi , and n. imoto , phys .
a * 59 * , 162 ( 1999 ) .
j. gruska , _ quantum computing _
( mcgraw - hill , new york , 1999 ) . | supersinglets @xmath0 are states of total spin zero of @xmath1 particles of @xmath2 levels .
some applications of the @xmath3 and @xmath4 states are described .
the @xmath3 states can be used to solve three problems which have no classical solution : the `` @xmath1 strangers , '' `` secret sharing , '' and `` liar detection '' problems . the @xmath4 ( with @xmath1 even )
states can be used to encode qubits in decoherence - free subspaces . |
in this work we continue our study of a system composed of an anderson impurity @xmath3 immersed in a 3d topological insulator ( 3dti ) with an inverted - mexican - hat " band dispersion around the @xmath4-point @xcite . for
the sake of self - consistency let us very briefly recapitulate the peculiar feature of the pertinent system : due to the special band structure , the ensuing kondo effect ( ke ) is profoundly distinct from its metallic analog in normal metals , because in addition to the original anderson impurity @xmath3 , there is an in - gap bound state ( henceforth denoted as an @xmath5 impurity ) , that is formed as a result of potential scattering@xcite .
the @xmath3 and @xmath5 impurities form a composite quantum impurity " ( cqi ) that turns the pertinent kondo physics much richer .
the reason for that is as follows : when isolated , the cqi can host two electrons that are found in singlet or triplet states with corresponding energies @xmath6 and @xmath7 . due to hybridization of the localized electrons in the cqi with the band electrons ,
the levels @xmath6 and @xmath7 are renormalized with decreasing bandwidth _ albeit with different rates _ @xcite . as a result ,
there is either a self - screened ke , or a ke with so(3 ) or so(4 ) dynamical symmetry ( depending on whether at the end of renormalization , @xmath7 lies above , below or coincides with @xmath6 @xcite . in our previous work@xcite ,
we have analyzed the pertinent kondo physics _ in the weak coupling regime _ using perturbative rg analysis techniques .
the goal of the present work is to perform an analysis of the kondo scattering with so(3 ) and so(4 ) dynamical symmetry in the _ strong coupling regime _ @xmath8 .
the main physical motivation is to elucidate the occurrence of singular fermi liquid in the so(3 ) symmetric sector , that is exposed _ only in the strong coupling regime_. more than a decade ago it has been shown that the traditional classification of quantum impurity models into fermi liquids and non fermi liquids should be modified in such a way that one has to distinguish between regular fermi liquids and singular fermi liquids ( sfl)@xcite . the former case is exemplified by the standard su(2 ) kondo model where electrons are scattered from a magnetic impurity of spin @xmath9 and at zero temperature the impurity is _
fully screened_. this makes it possible to describe the system in terms of nozires fermi - liquid picture . on the other hand ,
it was shown in ref .
@xcite that when the impurity is _ under - screened _ , the corresponding fermi liquid is singular .
practically , it implies that the density of states is ( logarithmically ) singular at the fermi energy . as is already stressed
, this manifestation of sfl occurs only at low temperature @xmath8 , and that requires the calculations to be carried out in the strong coupling regime , where perturbation theory is not applicable .
one need to resort to other approaches , such as bethe ansatz or conformal field theory ( cft , that is employed here ) . calculated experimental observables for so(3 ) and so(4 ) symmetric kondo effect include the impurity contribution to the temperature dependence of the specific heat , magnetic susceptibility and resistivity . comparing these results with those of the standard su(2 ) kondo effect
, we indeed elucidate the remarkable sfl nature of the ke for the so(3 ) symmetric sector .
experimentally , under - screened kondo effect ( uske ) has been observed in tunneling transport measurements through molecular transistors @xcite .
it is shown therein that the tunneling conductance @xmath10 approaches the unitary limit when the temperature @xmath0 tends to zero , but the derivative @xmath11 is divergent .
the possibility of detecting uske in kondo scattering in the bulk of metals is hampered by the fact that quantum impurities with high spin give rise to coqblin - schrieffer type of kondo scattering which eventually flows to a regular fermi liquid fixed point @xcite . in this work
we overcome this obstacle and demonstrate the feasibility of observing uske in bulk materials employing the occurrence of composite impurities .
the organization of the paper is facilitated by the fact that the model and the relevant starting kondo hamiltonian as well as the kondo temperature have already been derived in our previous work ( starting from the single impurity anderson hamiltonian ) .
therefore , in sec . [ sec - hk ]
we start right away by writing down the kondo hamiltonian @xmath12 describing low - energy exchange scattering of the band electrons by the @xmath3 and @xmath5 impurities as well as an exchange interactions between the @xmath3 and @xmath5 impurities .
the hamiltonian @xmath12 possesses different dynamical symmetries , so(3 ) or so(4 ) , within various energy domains .
the local density of states ( dos ) of the ti with magnetic impurity immersed in it is calculated in sec .
[ sec - ke - spin - s ] .
kondo interaction of the `` dressed '' fermi liquid ( formed by the band electrons and the @xmath3 impurity ) with the @xmath5-impurity is discussed in sec .
[ sec - second - tk ] . in sec .
[ sec - specific - heat ] we study and calculate the temperature dependence of the impurity contribution to the specific heat , while the magnetic susceptibility of the impurity is considered in sec .
[ sec - susceptibility ] .
[ sec - resist ] is devoted to the calculations of electric resistivity .
the results are then summarized in sec .
[ sec - conclusions ] .
some technical details are relegated to the appendices . in appendix [ append - model ] we describe the ground state of the isolated composite impurity .
weak coupling renormalization of the exchange interaction strength of the @xmath5-impurity with the band is considered in appendix [ append - scaling - so(4 ) ] .
as it is shown in our previous work @xcite , the lowest - energy states of the isolated cqi are the singlet and triplet states with one electron on the @xmath3-level and a second electron on the @xmath5-level .
electron tunneling between the @xmath3-level and the band modifies the number of electrons in the cqi .
integrating out high energy states from the band edges renormalizes the singlet and triplet levels until charge fluctuations are quenched and one arrives at the local moment regime . in this regime ,
the schrieffer - wolff transformation is used to map the anderson hamiltonian onto an effective hamiltonian @xmath13 . here
the first term , @xmath14 , describes electrons in the bulk of the ti , @xmath15 where @xmath16 denotes the conduction and valence band .
@xmath17 is the band dispersion .
[ subeqs - h - lead - def ] accordingly , @xmath18 is gapped and the insulator is topological for @xmath19 .
for @xmath20 ( assumed hereafter ) , the band dispersion has an `` inverted - mexican - hat '' form with dispersion minimum at a surface of nonzero wave - vector @xmath21 s , with @xmath22 where @xmath23 .
the kondo hamiltonian @xmath12 assumes the following form @xcite : @xmath24 where @xmath25 or @xmath26 is the localized spin of the @xmath3- or @xmath5-impurity , @xmath27 is the spin operator of the band electrons , @xmath28 @xmath29 is the vector of the pauli matrices . the couplings @xmath30 , @xmath31 and @xmath32 are explicitly given as @xcite , @xmath33 [ subeqs - j - def ] where @xmath34 here @xmath35 and @xmath36 are single electron energies of the @xmath3- and @xmath5-impurities , @xmath37 is hybridization rate of the @xmath3-impurity and the band and @xmath38 is the hybridization of the @xmath3- and @xmath5-impurities [ see fig .
[ fig - lead - d - f ] below ] , and @xmath39 is the coulomb blockade parameter for the @xmath5-impurity .
the coulomb blockade @xmath40 of the @xmath3-impurity is assume to be infinity @xcite .
notice that , generally , @xmath41 @xcite .
the low - energy physics of the model depends on @xmath32 .
more concretely , there are three different regimes for @xmath32 determining the different ground states ( gs ) @xcite : * when @xmath42 , the impurities are decoupled , and kondo scattering of the band electrons is determined mostly by the hybridization of the @xmath3-impurity with the band , whereas the @xmath5-impurity can be considered as an isolated magnetic moment ( i.e. , it is not coupled to the band or to the @xmath3-impurity ) . in this case
, there is standard kondo effect with full screening of the spin of the @xmath3-impurity .
the exchange interaction of the @xmath5-impurity with the band manifests itself just at temperature below kondo temperature .
therefore , @xmath31 enters into consideration only for temperatures below the kondo temperature .
* when @xmath43 with @xmath44 , the ground state of the cqi is a triplet and the singlet state is highly excited . in this case
, band electrons are scattered from a cqi with spin @xmath45 , and there is an uske of the magnetic impurity .
since the magnetic moments of the @xmath3- and @xmath5-impurities in the triplet ground state are parallel to one another , the exchange interaction strength @xmath31 of the @xmath5-impurity with the lead gives rise to slight modification of @xmath46 @xcite , ( that is the exchange interaction strength of the @xmath3 impurity with the band electrons ) .
therefore we assume @xmath47 in this regime .
* when @xmath48 with @xmath49 , the ground state of the cqi is singlet and the triplet state has high excitation energy . in this case , there is no kondo effect .
before presenting our calculations pertaining to physical observables in our system , let us recapitulate the nature of the gs , specific heat and magnetic susceptibility for the ordinary case of a magnetic impurity immersed in an _ ordinary metal _
[ ] . for the fully screened ( fs ) and the under screened ( us ) cases of the ke
they are listed in the table below .
recall that the gs of the fske is a fermi a liquid ( fl ) while that of the uske turns out to be a sfl .
explicitly , [ cols="^,^,^,^,^",options="header " , ] in the following , we will consider physical properties of the system in the strong coupling regime ( when the temperature is below the kondo temperature ) for the cases @xmath43 and @xmath42 , in turn .
in this section we apply the already known bethe ansatz expression for the scattering phase shift and write down the density of states ( dos ) that is peculiar to the system of spin s cqi immersed in a material with inverse mexican hat band structure .
specifically , we focus on the low - energy spectrum of the system where it is justified to approximate the dispersion ( [ disp ] ) by linear expressions in @xmath50 ( @xmath51 ) , where @xmath52 and @xmath53 are two solutions of the equation @xmath54 ( see fig . [ fig - disp ] ) . , eq .
( [ disp ] ) .
@xmath55 ( @xmath51 ) are two solutions of the equation @xmath54 .
@xmath56 are the fermi velocities at the inner or outer fermi surfaces ( @xmath57 or @xmath53).,width=60 ] the dos @xmath58 of the fermi gas near the impurity position is expressed in terms of the phase shift @xmath59 through the friedel sum rule @xcite , @xmath60 where @xmath9 or @xmath61 is the impurity spin .
calculations based on the bethe anzats @xcite applied for the full screened @xmath62 and the under - screened @xmath45 kondo effect , yield the following expression for the phase shift , @xmath63 for @xmath9 , the above expression takes the form , @xmath64 whereas for @xmath45 and @xmath65 , @xmath59 has the asymptotic expression , @xmath66 note that @xmath67 demonstrates a singular behavior near the point @xmath68 , whereas @xmath69 is regular .
differentiating @xmath70 ( [ delta - ba ] ) gives the following expression for the dos @xcite : @xmath71 .
\label{dos - bethe - anzats}\end{aligned}\ ] ] here @xmath72 or @xmath61 is the spin of the impurity , and the function @xmath73 is defined as @xmath74 where @xmath75 is the digamma function , @xmath76 ) for the so(4 ) ke ( blue curve ) and the so(3 ) ke ( red curve).,width=60 ] the dos eq .
( [ dos - bethe - anzats ] ) is shown in fig .
[ fig - dos ] for @xmath9 and @xmath45 .
it is seen that the dos for @xmath45 is singular . as a result the conventional fermi liquid expansion of the phase shift can not be carried out @xcite .
the origin of this singularity is the non - analytic behavior of the phase shift @xmath67 , eq .
( [ delta - ba ] ) , near the point @xmath68 .
when @xmath77 [ where @xmath78 is the kondo temperature for the so(4 ) ke , see ref .
@xcite and eq .
( [ tk4-def ] ) in appendix [ append - scaling - so(4 ) ] ] , the @xmath3-impurity and the conduction band electrons form a singlet state , and the system composed of the @xmath3-impurity _ and the band electrons _ forms a local fermi liquid that can be described within nozires fermi liquid theory@xcite .
the @xmath5-impurity is coupled to this local fermi liquid through an effective exchange hamiltonian , @xmath79 where @xmath80 the density of states of the local fermi liquid is , @xmath81 the dimensionless coupling @xmath82 is , @xmath83 where we take into account that @xmath84 .
the scaling equation for @xmath82 is , @xmath85 the initial value @xmath86 of @xmath87 is given by eq .
( [ tilde - jf - def ] ) .
the solution of the scaling equation ( [ eq - tilde - jf ] ) is , @xmath88 where the scaling invariant , the second kondo temperature , is @xmath89 for @xmath90 , @xmath91 , @xmath92 , @xmath93 and @xmath94 , we get @xmath95 and @xmath96 .
having set up the calculation framework we turn now to elucidate numerous physical observable .
these include specific heat , magnetic susceptibility and impurity resistivity .
since @xmath97 is very small , we shall restrict ourselves to the temperature regime @xmath98 in the following .
for the so(4 ) ke , the temperature is in the range @xmath99 . for the so(3 ) ke , we consider @xmath100 , where @xmath101 and @xmath78 are the kondo temperatures for the so(3 ) and so(4 ) ke , respectively @xcite . for the so(3 ) ke , we employ cft techniques , whereas for the so(4 ) ke , we will apply cft techniques for the interaction between the @xmath3-impurity and the conduction band electrons , and then we employ poor man s scaling formalism to take into account the interaction between the local fermi liquid and the @xmath5-impurity .
the first observable to be calculated is the specific heat of the ti within which the cqi is immersed .
its definition reads , @xmath102 where @xmath103 is the fermi function . noting that @xmath104 we have @xmath105 for the so(3 ) ke
, the specific heat is @xmath106 [ given by eq .
( [ specific - heat ] ) ] . for the so(4 ) ke , there is contribution to the specific heat due to the @xmath5-impurity .
this contribution can be written as , @xmath107 where @xmath108 is given by eq .
( [ tilde - jf - solution ] ) . the total specific heat for the so(4 )
ke is @xmath109 .
[ where @xmath110 is the specific heat ( [ specific - heat ] ) ] for the so(4 ) , so(3 ) and su(2 ) ke ( solid blue , solid red and dashed blue curves ) . here @xmath111 for the so(4 ) and su(2 ) ke or @xmath101 for the so(3 ) ke.,width=60 ] the ratio @xmath112 [ where @xmath110 is the specific heat ( [ specific - heat ] ) ] is shown in fig .
[ fig - cv - cvt ] for the so(4 ) and so(3 ) ke . in this case , the kondo temperature is @xmath78 . for comparison , the specific heat for the su(2 ) ke ( i.e. , for the case when the @xmath5-impurity is absent ) , is shown as well .
it is seen that for the su(2 ) ke , @xmath112 saturates to constant as @xmath2 , in agreement with the prediction of fermi liquid theory . the ratio @xmath112 for the so(4 ) ke is close to that for the su(2 ) ke for @xmath0 close to @xmath78 . for low temperatures , @xmath113 for the so(4 ) and su(2 ) ke deviate substantially from one another .
this is result is due to the exchange interaction of the @xmath5-impurity with the local fermi liquid . for the so(3 ) ke , the ratio @xmath112 diverges indicating the feature of a sfl behavior @xcite .
when @xmath43 and the ground state of the isolated cqi is @xmath45 , the magnetic susceptibility is , @xmath114 where @xmath115 is given by eq .
( [ dos - bethe - anzats ] ) with @xmath45 . taking into account that @xmath116 we can write , @xmath117 where @xmath118 for the case @xmath42 ( see eq .
( [ subeqs - j - def ] ) ) , the susceptibility is calculated in the following way : first , we employ the cft technique to calculate the susceptibility for @xmath119 ( see eq .
( [ tilde - jf - solution ] ) ) . within this approximation ,
the cqi splits into two noninteracting impurities , one of them is coupled to the band electrons , and the other is fully decoupled . in this case , the susceptibility is given by , @xmath120 where @xmath121 is given by eq .
( [ dos - bethe - anzats ] ) with @xmath9 .
the second term , @xmath122 , is the susceptibility of the @xmath5-impurity .
taking into account eq .
( [ df - de - def ] ) , we can write , @xmath123 where @xmath124 is given by eq . ( [ chi0-def ] ) .
the contribution to the susceptibility due to the interaction of the local fermi liquid with the @xmath5-impurity is , @xmath125 where @xmath108 is given by eq .
( [ tilde - jf - solution ] ) .
the total susceptibility for the so(4 ) ke is @xmath126 for the so(3 ) , su(2 ) and the so(4 ) ke is displayed as a function of temperature .
the saturation as @xmath2 for the su(2 ) and so(4 ) ke is reminiscent of the curie law . on the other hand , for the so(3 )
ke , @xmath127 . ,
title="fig:",width=60 ] [ fig - suscept - b ] the functions @xmath128 for the so(3 ) , su(2 ) and so(4 ) ke are shown in fig .
[ fig - suscept ] .
the su(2 ) ke corresponds to the case when the @xmath5-impurity is absent . for this case ,
the kondo temperature is @xmath78 .
it is seen that for the su(2 ) or so(4 ) ke , @xmath129 for @xmath130 .
for the so(3 ) ke , @xmath128 vanishes as @xmath131 when @xmath130 , which is a manifestation of a singular fermi liquid fixed point .
in this section we will calculate the impurity contribution to the resistivity . here
, the peculiar band structure of the ti plays a central role .
the calculation method of the impurity contribution to the resistivity depends on the underlying symmetry .
the cft technique used above to derive the specific heat and magnetic field will be employed to derive the resistivity of the ke with the so(3 ) symmetry . on the other hand , for the ke with the so(4 ) symmetry
, we apply a somewhat different approach : first , we apply the cft technique to derive the resistivity for the case @xmath119 ( see eq .
( [ tilde - jf - solution ] ) ) . in the next step ,
we apply the perturbative rg ( poor man s scaling approach ) to get the correction to the resistivity due to the interaction of the @xmath5-impurity with the local fermi liquid . for @xmath119
, the impurity resistivity can be written as @xcite @xmath132 here the conductivity @xmath133 is @xmath134 v_k^2 \tau_{\mathrm{tr}}(k ) , \label{conduct - def}\end{aligned}\ ] ] [ subeqs - resist - conduct ] where @xmath103 is the fermi - dirac distribution , and @xmath135 is the group velocity , @xmath136 for the dispersion ( [ disp ] ) , @xmath137 is given explicitly by , @xmath138 where @xmath139 the resistivity is directly related to the inverse of the transport relaxation time @xmath140 that is expressible in terms of the phase shifts @xmath141 ( corresponding to angular moment @xmath142 ) appearing in the partial wave expansion of the electron scattering from the impurity [ see eq .
( 2.20 ) in ref.@xcite ] .
explicitly , @xmath143 , \label{relax - time - phases}\end{aligned}\ ] ] where @xmath144 is the bare density of states ( dos ) , @xmath145 @xmath146 is the dirac delta function and @xmath147 is the linear size of the bulk . for the dispersion ( [ disp ] ) , the dos is @xmath148 where @xmath149 [ subeqs - dos - res ] @xmath51 . here
@xmath150 and @xmath151 correspond to the two solutions of the equation @xmath152 . for pure s wave scattering , @xmath153 and @xmath154 for @xmath155 .
( [ relax - time - phases ] ) can be written as , @xmath156 the scattering phase @xmath59 is given by eq .
( [ delta - ba ] ) . for @xmath157 , @xmath59 is given by eq .
( [ delta - s=1/2 ] ) for @xmath9 and by eq .
( [ delta - s=1-asymptot ] ) for @xmath45 .
* zero temperature resistivity * : as @xmath2 we can write @xmath158 substituting eq .
( [ df - t=0 ] ) into eq .
( [ conduct - def ] ) and taking into account eqs .
( [ group - velocity ] ) , ( [ delta - ba ] ) and ( [ subeqs - dos - res ] ) , we get @xmath159 when the fermi energy is constrained according to @xmath160 the equation @xmath54 has two solutions , @xmath57 and @xmath161 , ( see fig .
[ fig - disp ] ) .
the equation for the zero - temperature resistivity then takes the form , @xmath162 * finite temperature resistivity * : when the temperature is below the kondo temperature @xmath1 , the conductivity ( [ conduct - def ] ) can be written as , @xmath163 substitute into each of integrals a factor unity , + @xmath164=@xmath165 and taking into account that for @xmath166 or @xmath167 , the equation @xmath168 has a single solution , @xmath169 or @xmath170 ( see fig .
[ fig - disp ] for illustration ) , we can write @xmath171 where @xmath172 and the functions @xmath173 are given by eq .
( [ gi - def ] ) . taking into account eqs .
( [ relax - time - phase - delta ] ) and ( [ dos = dos1+dos2-res ] ) , we can write the conductivity ( [ conduct - relax ] ) as , @xmath174 the integrands on the right hand side of eq .
( [ conduct - phases - delta ] ) have a factor @xmath175 which is equal to 1 for @xmath176 and rapidly vanishes for @xmath177 .
the behaviour of the other factors , @xmath178 and @xmath179 ( @xmath51 ) , depends on the ratio @xmath180 .
when @xmath181 , then @xmath178 and @xmath179 change slowly within the interval @xmath182 and can be safely replaced by @xmath183 and @xmath184 . in what follows , we will assume the inequality @xmath185.@xcite then for @xmath8 the conductivity ( [ conduct - phases - delta ] ) takes the form , @xmath186 d\epsilon } { 4t\cosh^2(\frac{\epsilon}{2t})}. \label{conductivity - tneq0-res}\end{aligned}\ ] ] ( the limits of the integration can safely be changed to @xmath187 ) .
finally , the impurity resistivity is , @xmath188 where @xmath189 is the zero temperature impurity resistivity , see eq .
( [ conductivity - t=0-res ] ) .
* contribution of @xmath82 to the resistivity * : the impurity resistivity for the ke with the so(3 ) symmetry is given by eq .
( [ resist - tneq0-res ] ) . that for the ke with so(4 ) symmetry is contributed also by the interaction of the @xmath5-impurity with the local fermi liquid . in order to derive this contribution , we apply the perturbation poor man s scaling technique , @xmath190 where @xmath108 is given by eq .
( [ tilde - jf - solution ] ) .
the constant @xmath191 is , @xmath192 where @xmath193 @xmath55 are two solutions of the equation @xmath54 , see fig .
[ fig - disp ] .
hence , the resistivity for the so(4 ) ke reads , @xmath194 where @xmath195 is given by eq . (
[ resist - tneq0-res ] ) . ) as a function of temperature for the so(3 ) and su(2 ) symmetries ( solid red and dashed blue curves ) , and the resistivity ( [ resist - tneq0-so(4 ) ] ) for the so(4 ) ke ( solid blue curve).,width=60 ] the resistivity ( [ resist - tneq0-res ] ) as a function of temperature is shown in figure [ fig - resist - cft ] for the so(4 ) , so(3 ) and su(2 ) symmetries .
recall that as @xmath196 , the resistivity for the su(2 ) symmetry is given by , @xmath197 this behaviour is typical for the fermi liquid zero temperature fixed point .
deviation from eq .
( [ resist - so4-low - t ] ) in the curves for the su(2 ) and so(4 ) ke is due to the kondo interaction between the @xmath5-impurity and the local fermi liquid .
the temperature dependence of the resistivity for the so(3 ) symmetry is rather distinct . indeed , for @xmath196
, we can write @xmath198 the singular temperature dependence of the resistance is the result of the singular energy dependence of the scattering phase ( [ delta - ba ] ) .
the present work is motivated by the quest to elucidate the elusive sfl behavior within bulk materials .
this property , associated with uske , has so far been observed only in electron tunneling experiments through quantum dots@xcite , where it is demonstrated that the tunneling conductance @xmath10 approaches the unitary limit as @xmath2 , but the derivative @xmath11 diverges logarithmically .
elucidating sfl in metals based on the uske ( e.g iron immersed in cupper ) is not obvious because magnetic impurities with high spin give rise to coqblin - schrieffer type of scattering which falls on a regular fermi liquid fixed point @xcite .
here we achieved this goal by analyzing kondo scattering of electrons off an anderson impurity in a 3d topological insulator with an inverted mexican hat " band dispersion .
we used the fact that the interplay between the anderson impurity and its induced in - gap bound state results in self - screened kondo effect or in the kondo effect with so(4 ) or so(3 ) dynamical symmetries . using the conformal field theory technique , we have calculated the low temperature ( @xmath199 ) dependence of the specific heat , the magnetic susceptibility and the electric resistivity of the impurity for both screened and under - screened kondo effect .
physical properties of the impurity for the under - screened case demonstrate zero temperature singularity which corresponds to a singular fermi liquid phase .
in addition to the sfl behavior exposed here , this system exposes an interesting screening mechanism where the impurity @xmath5 is screened by the quasi - particles of the fermi sea that is formed when the band electrons and the @xmath3 impurity form a singlet state at @xmath8 .
* acknowledgements * : we acknowledge support by hkrgc through grant hkust03/crf09 . the research of i.k .
, t.k . and y.a .
is partially supported by grant 400/12 of the israel science foundation ( isf ) .
part of the material presented in this appendix has already been developed in our previous paper @xcite where we analyzed the same system in the weak coupling regime .
it is included here for the sake of self - consistency . in the first part
we write down the bare anderson model and specify the peculiarities resulting from the special form of the band structure . in the second part we elaborate on the scaling equations and rg flow for the so(4 ) dynamical symmetry .
the system considered here is schematically displayed in fig .
[ fig - lead - d - f]a .
it consists of a topological insulator whose energy band has the `` inverted mexican hat '' structure , and an anderson impurity immersed in its bulk ( denoted as @xmath3-impurity ) .
potential scattering of electrons on the impurity result in the formation a mid - gap localized energy level @xcite , denoted hereafter as an @xmath5-impurity .
the effective hamiltonian of the system is , @xmath200 here the first term , @xmath14 , is the hamiltonian ( [ h - ti - def ] ) describing electrons in the bulk of the ti . here
@xmath202 and @xmath203 are the ( atomic ) hamiltonians of the @xmath3- and @xmath5-impurities , and @xmath204 describes the hybridization between them , @xmath205 where @xmath35 or @xmath36 is the @xmath3- or @xmath5-impurity energy level and @xmath40 or @xmath39 is the interaction between electrons on the impurity . here
@xmath206 , @xmath207 , @xmath208 or @xmath209 is the creation or annihilation operator of electron on the @xmath3-level , @xmath210 or @xmath211 is the creation or annihilation operator of electron on the @xmath5-level .
[ subeqs - h - complex - impurity - def ] the last term on the right hand side of eq .
( [ h - tot - def ] ) , @xmath212 , is the hybridization between the anderson impurity @xmath3 and the band electrons , @xmath213 note that hybridization prevails only between the band electrons and the @xmath3 impurity .
the @xmath5-level is formed as a result of the potential scatering of the conduction band electron on the @xmath3-impurity @xcite .
therefore , hybridization of @xmath5-impurity and the band electrons is absent .
* energy scales * : few words about energy scales are in order : unless otherwise specified , we shall assume that @xmath214 and @xmath215 where @xmath216 ( the initial bandwidth ) is the highest energy cutoff , and @xmath217 is the fermi energy ( see figure [ fig - lead - d - f ] ) .
in the calculations below we use @xmath92 , @xmath218 , @xmath219 and @xmath220 . *
the eigenstates of @xmath221 * , equation ( [ hc - def ] ) , are specified by the configuration numbers @xmath222 indicating the number of electrons on the levels @xmath3 and @xmath5 . with energy scales specified in eq .
( [ escale ] ) , the ground state ( gs ) has @xmath223 and there are four possible states , a spin - singlet state @xmath224 and three spin - triplet states @xmath225 ( @xmath226 . the singlet energy is modified when @xmath227 while the triplet energy is unaffected .
explicitly , in order to perform the rg analysis considered in sec .
[ sec - second - tk ] [ see eq .
( [ eq - tilde - jf ] ) ] , we need the effective coupling @xmath232 at the effective bandwidth equal to @xmath78 . to derive @xmath232 , we apply the weak coupling rg analysis for the so(4 ) ke . in the the weak coupling regime ,
the scaling equations for the dimensionless couplings @xmath233 and @xmath234 must proceed to third order in these parameters @xcite , that is , scaling of @xmath236 and @xmath237 is shown in the flow diagram , fig .
[ fig - scaling ] .
the red dot denotes the two - channel fixed point @xmath241 .
we are interested in the solution of these equations under the following inequalities : @xmath242 so that we are far away from the two - channel fixed point . in this case
@xmath243 increases when @xmath244 decreases , whereas @xmath245 decreases and goes to zero when @xmath244 vanishes .
the solution of the scaling equation ( [ eq - jd ] ) can be approximated as , @xmath246 where the scaling invariant , the kondo temperature @xmath78 , is @xmath247 the solution ( [ jd - solution - scaling - limit ] ) is valid just for @xmath248 .
a more satisfactory approximation is given in ref .
@xcite , @xmath249 the solution of the scaling equation ( [ eq - jf ] ) is , @xmath250 taking into account eq .
( [ jd - solution - scaling - limit - beter ] ) , we get @xmath251 where @xmath252 when @xmath244 approaches @xmath78 , @xmath232 can be approximated as @xmath253 where we take into account that @xmath254
. j. j. parks , a. r. champagne , t. a. costi , w. w. shum , a. n. pasupathy , e. neuscamman , s. flores - torres , p. s. cornaglia , a. a. aligia , c. a. balseiro , g. k .- l .
chan , h. d. abrua , d. c. ralph , science * 328 * , 1370 ( 2010 ) ; arxiv:1005.0621 . | electron scattering off an anderson impurity immersed in the bulk of a 3d topological insulator is studied in the strong coupling regime , where the temperature @xmath0 is lower than the kondo temperature @xmath1 .
the system displays either a self - screened kondo effect , or a kondo effect with so(3 ) or so(4 ) dynamical symmetries .
low temperature kondo scattering for systems with so(3 ) symmetry displays the behavior of a singular fermi liquid , an elusive property that so far has been observed only in tunneling experiments .
this is demonstrated through the singular behavior as @xmath2 of the specific heat , magnetic susceptibility and impurity resistivity , that are calculated using well known ( slightly adapted ) conformal field theory techniques .
quite generally , the low temperature dependence of some of these observables displays a remarkable distinction between the so(n=3,4 ) kondo effect , compared with the standard su(2 ) one . |
in 1964 gell - mann and zweig independently proposed a quark model in which all known hadrons were described as bound states of only three fundamental particles @xcite .
each of these spin-1/2 particles , which gell - mann named quarks , has their corresponding antiparticles . however , a precise mechanism of the bound states and quark confinement has still not been established .
gell - mann s model requires that the valid quark states should be colorless and so a free quark is not allowed . since a free single quark has not yet been observed , the model assumes that quarks appear as bound states .
the baryons are bound states of three quarks ( or antiquarks ) and the mesons are made up of one quark and one antiquark . as a model , to describe this behavior , one can think of a hadron as a cavity in which the quark wave function is confined , and the cavity is surrounded by the qcd vacuum @xcite . therefore in a simple model quarks confined in a sphere with radius @xmath6 and the dirac equation
sould be solved within this sphere . from this point of view
, the solution of the dirac equation in a three - dimensional scalar potential , apart from its interesting theoretical aspects , provide useful tools for studying the properties of elementary particles .
for example , the mit bag model @xcite and its charily invariant versions , such as the chiral bag model @xcite and the cloudy bag model @xcite , are some models for describing the physics of the nucleon and other baryons .
a bag is a region of space in which quarks and gluons are confined , i.e. they are forced by an external constant pressure b , which can be fitted using experimentally determined hadron masses , to move only inside the bag .
historically , chodos _ et .
have considered the mit bag model through the dirac equation ( except the bag pressure b ) @xcite .
they solved bag equations for the massless dirac fields in three space dimensions .
their solutions are for the special case of static spherical boundary .
they also computed charge radius and found it to be 1.0 fm .
al . _ , in the other case of this model , calculated the masses and the static parameters of the light hadrons @xcite . in the cloudy bag model a baryon is treated as a three - quark bag that is surrounded by a cloud of pions .
thomas _ et . al .
_ investigated the static properties of the nucleon within this model @xcite .
they found the bag radius to be about 0.8 fm by a fit to pion - nucleon scattering in the ( 3,3)-resonance region .
the chiral bag model for the nucleon is a hybrid of quark and meson degrees of freedom , interpolating the two limits of the skyrme model at @xmath7 and the mit bag model at @xmath8 @xcite . in the skyrme model mesons acts as gauge particles so that baryons would interact with each other by the exchange of mesons @xcite .
skyrmions are the solutions of the field equations
. these solutions are solitons and no longer plain wave .
one may interpret these skyrmions as coherent states of baryons and excited baryons @xcite ; however , the physical interpretation is still not completely resolved . with this model ,
it is also possible to calculate nucleon masses and other particle properties @xcite .
both the mit bag and skyrme model are useful to calculate masses and other properties of hadrons . in ref .
@xcite , hosaka and toki investigated the static properties of the nucleon such as masses and magnetic moments as a function of @xmath6 , in both the original chiral bag model and models with vector mesons .
the mit bag model introduces many free parameters for energy corrections that could be helpful in understanding the physical processes inside the nucleus .
it is a curious and complex situation to solve the relativistic quantum mechanics problems in a finite potential in comparison with the equivalent problem in nonrelativistic quantum mechanics . in the dirac equation , the wave function is continuous , its first derivative is discontinuous and the second derivative has a very large jump , whereas , in the schrdinger equation the wave function and its first derivative are continuous , but the second derivative has a certain jump related to the potential jump .
the solutions of the dirac equation in a @xmath9 potential exist in the literature , see for example @xcite .
however , for a finite spherical potential well , as far as we are aware of , there is no solution in the literature . in this paper
, we consider the solution of the dirac equation in a spherically symmetric scalar potential well , which is not necessarily infinite .
the origin of this scalar potential could be a strong force that binds quarks together in clusters to make more familiar subatomic particles , such as protons and neutrons .
it also holds together the atomic nucleus and underlies interactions between all particles containing quarks @xcite .
although this model is not compatible with the quark confinement at finite potential , it could serve a dual purpose .
first , one could observe the evolution of the wave functions as @xmath10 , where one could recover the mit bag model results .
second , for finite values of @xmath11 this could serve as an effective model for the nucleus where @xmath5 is the effective single particle potential emerging from the meson exchange of the nucleons . here
we obtain eigenvalue equations for the energies and numerically calculate the energy eigenvalues for the @xmath1 and @xmath3 states .
these states are derived normally from the continuity of the wave function and imposing boundary conditions on the cavity @xcite , with different radii @xmath12 fm and the quark masses @xmath13 @xmath14 .
we then , compare the values of the energy levels in each of specific conditions and also with the energy eigenvalue obtained in the previous mit bag model .
the relation between energy eigenvalues and the radius of cavity and the mass of the quark is considered .
we also obtain the dirac wave function components for a quark particle and depict them in figures . as a result ,
when the strength of the potential is increased , the wave function components on the boundary of cavity fall down and it would have the sharper point , which are in good agreement with the mit bag model . finally , we obtain the magnitude of sharpness of the relativistic wave function component when crossing the wall . the paper is organized as follows : section [ sec2 ] is devoted to an introduction of the dirac equation with a central scalar potential . in sec
[ sec3 ] we calculate the energy eigenvalues for a dirac particle in a scalar potential with full spherical symmetry for the @xmath1 and @xmath3 states .
then we discuss about the sharpness of the wave function components on the boundary
. finally in sec [ sec4 ] we summarize our results .
in the dirac equation , the scalar potential @xmath15 and the fourth component of a vector potential , @xmath16 are accompanied by mass @xmath17 and energy @xmath18 , respectively .
although @xmath16 is not a vector potential , since it is the fourth component of a four vector , it is called a vector potential .
the general form of the equation of motion for a spin-@xmath19 particle with these two potential is ( in relativistic units , @xmath20 and @xmath21 ) @xmath22\psi ( x , t ) = i\frac{\partial } { { \partial t}}\psi ( x , t),\ ] ] where @xmath23 is the three - dimensional momentum operator . in the above equation @xmath24 and @xmath25
are the @xmath26 dirac matrices which , in the usual representation , are given by @xmath27 where @xmath28 is the @xmath29 unit matrix .
the subscript @xmath30 can take the values of 1,2,3 , and @xmath31 are the @xmath29 pauli matrices . for a dirac particle in a spherically symmetric potential field , the total angular momentum operator @xmath32 , and the spin - orbit matrix operator @xmath33 , commute with the dirac hamiltonian . here
@xmath34 is the orbital angular momentum .
the complete set of the conservative quantities with their eigenvalues can be written as follows : @xmath35 so that , @xmath36 = 0,\qquad\left [ { h,{j } } \right ] = 0,\qquad \left [ { j^2 , j_3 } \right ] = 0,\qquad\left [ { { j},k } \right ] = 0,\ ] ] @xmath37 therefor the quantum number @xmath38 is a nonzero integer number .
given stationary solutions @xmath39 , we have @xmath40 where @xmath41 and @xmath42 are real square - integrable functions , and @xmath43 and @xmath44 can be written in terms of the spherical harmonic functions with the relevant clebsch - gordan coefficients .
then the two coupled equations for the radial parts of the dirac equation with a given scalar potential turn out to be , @xmath45 the solutions of eqs .
( [ 7 ] ) and ( [ 7.1 ] ) for a scalar potential well , @xmath46 and @xmath47 , are the spherical bessel functions and modified spherical bessel functions of the first kind , for the regions @xmath48 and @xmath49 , respectively
. therefore for region i , @xmath50 with @xmath51 , and we have [ 8 ] @xmath52 and for region ii , @xmath49 with @xmath53 , [ 10 ] @xmath54 where @xmath55 and @xmath56 , and @xmath57 , and @xmath58 are the normalization factors @xcite .
in this section we compute the energy eigenvalues for a dirac particle in a full spherically symmetric scalar potential .
reported eigenvalues , for which the massless spin-@xmath19 field is confined to an infinite spherical potential well of radius @xmath59 fm , are listed in table [ table1 ] .
these values have been computed through an equation of motion and boundary conditions @xcite . | in this paper we investigate a solution of the dirac equation for a spin-@xmath0 particle in a scalar potential well with full spherical symmetry .
the energy eigenvalues for the quark particle in @xmath1 states ( with @xmath2 ) and @xmath3 states ( with @xmath4 ) are calculated .
we also study the continuous dirac wave function for a quark in such a potential , which is not necessarily infinite .
our results , at infinite limit , are in good agreement with the mit bag model .
we make some remarks about the sharpness value of the wave function on the wall .
this model , for finite values of potential , also could serve as an effective model for the nucleus where @xmath5 is the effective single particle potential . : 12.39.-x , 03.65.pm |
spurious two - level system ( tls ) fluctuators are considered a serious source of low - frequency noise in superconducting qubits @xcite , and the characterization of these fluctuators in solid - state devices has a long history @xcite .
most recently , coherent coupling between tls fluctuators and a superconducting phase qubit was observed via the novel energy splittings in spectroscopic measurements @xcite .
it was shown that the tls fluctuators have much longer decoherence times than the superconducting qubits , raising the possibility of realizing quantum manipulation on these fluctuators @xcite .
the key question in manipulating the tls fluctuators is how to implement the required coherent manipulation and readout .
located sparsely inside solid - state devices , the fluctuators usually do not interact with each other , and their states are hard to control .
the coupling between the fluctuators and solid - state devices provides us with a tool to achieve the quantum manipulation @xcite .
however , conventional gate schems using cavity qed approach that are usually exploited for solid - state qubits can not be applied to this system because of the large energy separation between the fluctuators . in this work
, we will present a gate scheme that exploits the controllability of the superconducting resonator to implement high fidelity gates on the tls fluctuators @xcite , even when the decay of the resonator is a few megahertz .
the superconducting resonator acts as a knob that controls the dynamics of individual fluctuator , as well as coupling them together . working with practical parameters from the superconducting josephson junction resonator , we will design single - qubit and two - qubit quantum logic gates in the presence of resonator decay .
our scheme takes into account the full coupling hamiltonian between the tls fluctuators and the resonator .
readout of the fluctuators can also be performed by measuring the transmission through the resonator .
this scheme can be extended to fluctuators in different josephson junctions by connecting the junctions into the same superconducting loop due to the nonlocal nature of the microwave mode of the resonator .
this work hence provides a realizable design for coherent manipulation of multiple tls fluctuators , which is closely related to current experimental efforts in studying the fluctuators and their coupling with superconducting resonator modes .
various superconducting resonators in the microwave regime , including superconducting transmission lines , josephson junctions , squid s , and superconducting lumped element resonators , have recently been demonstrated and have shown quantum behavior and strong coupling with superconducting qubits @xcite .
superconducting resonators are also promising systems for studying quantum effects such as single photon generation and lasing @xcite , and one of us has shown recently that a josephson junction can be used to probe various properties of tls fluctuators , e.g. to resolve the mechanism that couples the fluctuators to the junction @xcite . while we will focus on the josephson junction resonator , we want to emphasize that our results can be readily generalized to other superconducting resonators @xcite .
the paper is organized as the following . in sec .
[ system ] , we will study the coupled system of the fluctuators and a josephson junction resonator , including the driving on the resonator . in sec .
[ dispersive ] , we will derive the effective hamiltonian for the tls fluctuators in the dispersive regime where the quantum operations are implemented .
we will also derive the residual coupling between the fluctuators and the resonator in this regime . in sec .
[ gates ] , we will present detailed scheme for single - qubit and two - qubit quantum logic gates . then , we will estimate the decoherence of the fluctuators during the gate operations in sec .
[ decoherence ] .
we will also test the fidelity of the quantum operations with numerical simulation of the full hamiltonian , taking the resonator decay into account . in sec .
[ discussions ] , we will discuss the readout of the fluctuators and the extension of gate scheme to fluctuators inside different junctions .
the conclusions will be given in sec .
[ conclusions ] .
consider the system in fig .
[ fig1 ] ( a ) , where tls fluctuators inside the amorphous layer of a josephson junction couple with the junction resonator in an rf squid loop . with total capacitance @xmath0 ,
josephson energy @xmath1 , loop inductance @xmath2 , and magnetic flux @xmath3 inside the squid loop , the hamiltonian of the resonator can be written as @xmath4 in terms of the phase @xmath5 and the conjugate momentum @xmath6 .
this hamiltonian can be approximated as an oscillator mode with a phase shift @xmath7 from the origin : @xmath8 and the phase shift satisfies @xmath9 the frequency of the resonator can be written as @xcite @xmath10 which can be tuned in a large range by the magnetic flux @xmath3 .
in addition , driving can be applied to the resonator by e.g. applying an external radio - frequency current @xmath11 to the resonator with @xmath12 .
a josephson junction resonator containing spurious two - level system fluctuators denoted by arrows .
( a ) fluctuators in a single junction and ( b ) fluctuators in different junctions . , width=302 ] the tls fluctuators reside inside the tunneling layer and can couple with the junction resonator by various mechanisms . for example , the coupling to the critical current of the junction takes the form @xmath13 , where @xmath14 is the polarization and magnitude of the coupling . denoting the resonator annihilation operator by @xmath15 with @xmath16 .
let @xmath17 for simplicity and @xmath18 be the driving frequency .
the total hamiltonian of the coupled system in the rotating frame can be written as @xmath19 \\
h_{\kappa } & = & \sum_{k}\hbar\omega_{k}a_{k}^{\dagger}a_{k}+c_{k}(a_{k}^{\dagger}a+a^{\dagger}a_{k } ) \end{aligned}\ ] ] where @xmath20 is the hamiltonian of the driven resonator mode with the detuning @xmath21 and the driving amplitude @xmath22 , @xmath23 is the hamiltonian of the fluctuators including the coupling between the fluctuators and the resonator mode , and @xmath24 is the hamiltonian of the thermal bath connected to the resonator . here , the index @xmath25 labels different fluctuators , @xmath26 are the pauli operators , @xmath27 is the detuning of the fluctuators , and @xmath28 is the coupling constant .
note that coupling constant for other coupling mechanisms such as dielectric coupling between the fluctuators and the resonator can be derived similarly @xcite .
the decay of the resonator is modeled by its coupling to a bath of modes described by the annihilation operator @xmath29 with frequency @xmath30 and coupling constants @xmath31 .
the decay rate is given by @xmath32 @xcite .
the hamiltonian @xmath33 describes a typical cavity qed system between the fluctuators and the junction resonator @xcite .
note that the driving on the resonator generates a time - dependent oscillation in the phase variable with the amplitude @xmath34 . to keep the nonlinear term in the josephson energy to be small
, the oscillation amplitude needs to be small , e.g. @xmath35 . with @xmath36 and typical parameters @xmath37 and @xmath38
, we estimate that the driving amplitude is bounded by @xmath39 .
in this work , we study the quantum logic operations in the dispersive regime where the coupling @xmath40 is much weaker than the detuning between the fluctuators and the resonator : @xmath41 with @xmath42 . in this regime , we can apply the following unitary transformation @xcite @xmath43 to the system .
after the transformation , the hamiltonian becomes @xmath44 with @xmath45 to the second order of @xmath46 .
the hamiltonian is now divided into three parts : a hamiltonian for the resonator @xmath47 , an effective hamiltonian for the fluctuators @xmath48 , and a small residual coupling between the fluctuators and the resonator @xmath49 .
the hamiltonian @xmath48 can be written as @xmath50 + h_{int}+\widetilde{h}_{k } \label{eq : h1eff}\\ h_{int } & = & \sum\lambda_{mn}(\sigma_{n+}\sigma_{m-}+\sigma_{m+}\sigma_{n-})/2\label{hint } \\
\widetilde{h}_{\kappa } & = & \sum_{n , k } ( g_{n}c_{k}/\delta_{nc})(\sigma_{n+}a_{k}+a_{k}^{\dagger}\sigma_{n-})\end{aligned}\ ] ] which includes the effective single qubit terms , an exchange - like interaction @xmath51 , and an induced coupling to the bath modes of the resonator @xmath52 .
we derive the detuning for the single qubits as @xmath53 and the rabi frequency as @xmath54 the coupling constant in the exchange - like interaction can be derived as @xmath55 in the following section , we will study the implementation of the quantum logic gates with the hamiltonian @xmath48 .
the residual coupling @xmath49 can be written as @xmath56 \label{eq : hx}\ ] ] where the first term is the stark shift for the resonator and the second term is a coupling to the resonator amplitude originated from the finite driving amplitude . because of the amplitude shift in the unitary transformation in eq .
( [ u ] ) , the average occupation of the resonator is now zero with @xmath57 .
hence , the first term has a small effect on the fluctuators during the quantum operations .
the second term can induce a small modification to the coupling constant @xmath58 in the effective interaction in eq .
( [ hint ] ) which will be studied in detail below .
universal quantum gates can be performed by controlling the effective hamiltonian @xmath48 .
here , we present the scheme for the single - qubit and two - qubit gates with typical parameters from superconducting josephson junction resonators .
upper plot : gate times for hadamard gate ( solid curve ) and spin flip gate ( dashed curve ) versus detuning @xmath59 .
lower plot : effective couplings @xmath60 and @xmath61 versus @xmath59 ( main plot ) ; energy @xmath62 ( solid curve ) and @xmath63 ( dashed curve ) versus @xmath64 at @xmath65 ( inset ) . ,
width=302 ] single - qubit gates on a chosen tls ( e.g. @xmath66 ) can be performed by adjusting the frequency of the driving source to be close to the frequency of this tls . by adjusting the driving amplitude and the detuning of the resonator ,
the effective qubit parameters @xmath67 and @xmath68 can be adjusted in a wide range . for a given detuning @xmath59
, one can adjust the driving amplitude @xmath64 in eq .
( [ delta ] ) to have @xmath69 and obtain the spin flip gate @xmath70 .
the gate time can be found to be @xmath71 at the chosen @xmath64 .
one can also adjust @xmath64 to have @xmath72 to implement the hadamard gate @xmath73 . here , the driving amplitude is chosen to be @xmath74 with the gate time @xmath75 . in fig .
[ fig2 ] , we plot the gate times of the hadamard gate and spin flip gate with the parameters @xmath76 and @xmath77 . it can be shown that gate times on the order of @xmath78 can be achieved .
.[tab1]example parameters for implementing the spin flip gate ( x ) and the hadamard gate ( h ) . [
cols="^,^,^,^,^",options="header " , ] one can estimate the gate fidelity approximately as @xmath79 which depends on the ratio between the gate time and the decoherence time . using the parameters given above , we can estimate the ratio @xmath80 . in table
[ tab2 ] , it is shown that @xmath81 for the two - qubit gate in our scheme and @xmath82 for the swap operation in the cirac - zoller gate at the damping rate @xmath83 .
the gate fidelities can thus reach @xmath84 with our protocols .
this indicates that the josephson junction resonator is an effective tool in generating high fidelity quantum operations even at a quality factor ( @xmath85 ) of only @xmath86 and hence can be used to demonstrate quantum coherence in the tls fluctuators . the term @xmath87 in the residual coupling @xmath49 in eq .
( [ eq : hx ] ) can induce additional decoherence . here , quantum fluctuations of the resonator mode cause extra noise even at zero resonator amplitude .
the spectrum of the quantum fluctuations can be derived as @xmath88 from which the decoherence rate can be derived as @xmath89 as can be seen , the decoherence rate is to the @xmath90th order of @xmath46 when the driving amplitude @xmath64 is comparable with the detunings . with the above parameters , we find that @xmath91 which can be neglected during the gate operations .
gate fidelity versus damping rate @xmath92 by numerical simulation .
solid curve : swap gate and dash - dot curve : hadamard gate .
two fluctuators are included in the simulation for both gates .
for the hadamard gate , we use the parameters @xmath93 and @xmath94 for tls @xmath95 . , width=302 ] to study the effectiveness of the quantum logic gates , we perform numerical simulations on the gate operations using the full hamiltonian @xmath96 . the decay of the resonator is simulated using the lindblad master equation @xcite .
this simulation includes both the effect of the residual coupling in eq .
( [ eq : hx ] ) and the effect of the resonator decoherence .
we calculate the fidelities of the hadamard and swap gates using the method prescribed by nielsen s formula for the gate fidelity @xcite over a wide range of resonator decay rate .
the results are plotted in fig .
this simulation shows that the fidelity can be higher than @xmath97 for @xmath98 for single - qubit and two - qubits gates , which also agrees with our estimations above .
hence , at moderate decay rate for the josephson junction resonator , high fidelity quantum logic gates can be achieved .
as demonstrated in recent works , the superconducting resonator can be used to detect the states of qubits or fluctuators @xcite . in the strong damping regime ,
the amplitude of the superconducting resonator adiabatically follows the dynamics of the tls fluctuators , as was shown in our previous work @xcite .
a phase sensitive measurement of the resonator can hence provide a direct measurement of the tls states . in the moderate damping regime where the damping rate is weaker than the coupling constant
, a measurement of a tls can be performed by adjusting the resonator frequency to be in the vicinity of this tls , but with the condition @xmath41 . here
, the stark shift resulting from this tls is much stronger than that from other qubits .
a measurement of the resonances in the transmission spectrum of the resonator can then provide a measurement of the qubit state @xcite .
such measurements can be realized with current electronics where the resonator and the driving can be adjusted and switched on - and - off in nanoseconds , much faster than the decoherence time of the fluctuators .
our scheme can also be extended to tls fluctuators inside different junctions . because the wavelength of the microwave mode of the superconducting resonator is much longer than the dimension of this circuit , fluctuators in different junctions can couple to the same resonator mode when the junctions are connected by a superconducting loop . as is illustrated in fig .
[ fig1 ] ( b ) , two junctions are connected to the central superconducting island labelled by @xmath5 that is associated with the junction resonator . it can be shown that effective coupling between fluctuators in the two junctions can be derived exactly as described by eq .
( [ eq : h1eff ] ) . in this configuration ,
quantum logic gates can be performed with essentially the same approach as was presented above .
this circuit can also be extended to include multiple junctions .
this system is thus intrinsically `` scalable '' where fluctuators in multiple junctions couple nonlocally .
note that the frequency of the superconducting resonator is determined by the total capacitance , the total effective josephson energy , and the inductance in the circuit , and will decrease as the number of junctions increases .
this can set a limit on the number of junctions that can be connected into the circuit .
tls fluctuators have been studied for a long time .
previously , the fluctuators are often considered as a source of decoherence in superconducting qubits , causing the so - called @xmath99 noise . in this work , we focused on studying the coherent manipulation of the fluctuators which can provide insights about the dynamics , the coupling mechanism , and the relaxation of the fluctuators in superconducting devices .
although the success in implementing universal quantum logic gates makes the tls fluctuators potential candidate for quantum computing , we want to emphasize that the main aim of this work is to provide a practical scheme to demonstrate the coherence behavior of the fluctuators .
our scheme can be useful for current experiments that investigate the coupling between the fluctuators and superconducting resonators @xcite .
to conclude , we have shown that universal quantum logic gates can be implemented on spurious tls fluctuators via the coupling between the fluctuators and the josephson junction resonator .
taking into account the full hamiltonian of the coupled system and the effect of the noise , our numerical simulation of the quantum operations showed that quantum logic gates can be performed with high fidelity even at resonator decay rates of a few megahertz .
we have used practical parameters for the junction resonators and the fluctuators in this study .
our work hence indicates that quantum coherence and quantum manipulation of tls fluctuators can be readily demonstrated .
the results here can be generalized to other types of superconducting resonators which are explored in recent experiments @xcite .
99 y. makhlin , g. schn , and a. shnirman , rev .
phys . * 73 * , 357 ( 2001 ) ; j. q. you and f. nori , phys . today * 58 * 11 , 42 ( 2005 ) ; m. h. devoret and j. m. martinis , quantum inf . process .
* 3 * , 163 ( 2004 ) .
p. dutta and p. m. horn , rev .
* 53 * , 497 ( 1981 ) ; m. b. weissman , rev .
phys . * 60 * , 537 ( 1988 ) ; f. c. wellstood , c. urbina , and j. clarke , appl .
* 85 * , 5296 ( 2004 ) ; d. j. van harlingen and _ et al .
b * 70 * , 064517 ( 2004 ) ; o. astafiev and _ et al . _ ,
lett . * 96 * , 137001 ( 2006 ) ; h. paik and _ et al .
b * 77 * , 214510 ( 2008 ) .
z. kim and _ et al .
_ , arxiv:0810.2455 ; b. s. palmer and _ et al .
b * 76 * , 054501 ( 2007 ) ; s. pottorf , v. patel , and j. e. lukens , arxiv:0809.3272 ; j. a. schreier and _ et al . _ ,
b , 180502(r ) ( 2008 ) ; y. yu and _ et al .
_ , arxiv:0807.0766 .
i. martin , l. bulaevskii , and a. shnirman , phys .
lett . * 95 * , 127002 ( 2005 ) ; a. shnirman , g. schn , i. martin , and y. makhlin , phys . rev .
lett . * 94 * , 127002 ( 2005 ) ; r. h. koch , d. p. divincenzo , and j. clarke , phys .
lett . * 98 * , 267003 ( 2007 ) ; m. constantin and c. c. yu , phys
99 , 207001 ( 2007 ) . | two - level system fluctuators in superconducting devices have demonstrated coherent coupling with superconducting qubits . here , we show that universal quantum logic gates can be realized in these two - level systems solely by tuning a superconducting resonator in which they are imbedded .
because of the large energy separation between the fluctuators , conventional gate schemes in the cavity qed approach that are widely used for solid - state qubits can not be directly applied to the fluctuators .
we study a scheme to perform the gate operations by exploiting the controllability of the superconducting resonator with realistic parameters .
numerical simulation that takes into account the decay of the resonator mode shows that the quantum logic gates can be realized with high fidelity at moderate resonator decay rate .
the quantum logic gates can also be realized between fluctuators inside different josephson junctions that are connected by a superconducting loop .
our scheme can be applied to explore the coupling between two - level system fluctuators and superconducting resonators as well as the coherent properties of the fluctuators . |
massive stars have a strong impact on the evolution of galaxies .
o type stars and their descendants , the wolf rayet stars , are the main source of uv photons , mass , energy and momentum to the interstellar medium .
they play the main role in the ionization of the interstellar medium and dust heating .
the milky way is the best place to access , simultaneously , massive stellar populations and their impact on the surrounding gas and dust .
the sun s position in the galactic plane , however , produces a heavy obscuration in the optical window ( @xmath5 mag ) toward the inner galaxy , where massive star formation activity is the greatest .
shifting to longer wavelengths , the perspective is much better , especially in the near infrared , because these wavelengths are long enough to lessen the effect of interstellar extinction ( @xmath6 mag ) and still short enough to probe the stellar photospheric features of massive stars @xcite .
the study of giant h ii regions ( gh ii ) in the near infrared can address important astrophysical questions such as : 1 . characterizing the stellar content by deriving the initial mass function ( imf ) , star formation rate and age ; 2 .
determining the physical processes involved in the formation of massive stars , through the identification of ob stars in very early evolutionary stages , such as embedded young stellar objects ( ysos ) and ultra compact h ii regions ( uch ii ) ; and 3 . tracing the spiral arms of the galaxy by measuring spectroscopic parallaxes of zero age main sequence ob stars .
the exploration of the stellar content of obscured galactic gh ii regions has been studied recently by several groups : @xcite , @xcite , @xcite . in particular , blum and collaborators presented near infrared imaging and spectroscopic observations of three optically obscured gh ii regions : w43 , w42 and w31 .
these observations revealed massive star clusters at the center of the h ii regions which had been previously discovered and studied only at longer wavelengths . in this work ,
we present results for ngc3576 ( g291.3 - 0.71 ) , located at a kinematic distance @xmath7 kpc , which we adopted from @xcite , after correcting for the standard galactic center distance ( @xmath8 kpc ) .
ngc3576 appears in the visible passbands as a faint h ii region , but in the infrared it is among the most luminous in our galaxy @xcite .
in fact , with 1.6 x 10@xmath9photons s@xmath10 inferred from the radio data , it can be classified as a gh ii ( following the suggestion of r. kennicutt for sources brighter than @xmath11 lyman continuum , @xmath12 @xmath13 , photons per second , private communication ) .
a gh ii region has at least ten times the luminosity of the orion nebula and roughly the number emitted from the hottest single o3-type star , thus implying multiple hot stars .
ngc3576 was observed in the radio continuum by @xcite .
@xcite , @xcite , @xcite , and @xcite have detected radio recombination lines .
maser sources have also been detected in the region : ch@xmath14oh @xcite and h@xmath15o @xcite .
the detection of intense emission in the 10 @xmath2 m window , h@xmath15o masers , and the compact thermal emission in the radio are typical indications of the primitive stages of star formation and of a dense circumstellar environment .
photometry from 1 to 2.5 @xmath2 m of the brightest sources was performed by moorwood & salinari ( 1981 ) and by @xcite showing that the spectral energy distributions of these objects suggests that they have excess emission .
an intense co j=21 line at 230 ghz was observed by @xcite in the core region of ngc3576 . in the present paper ,
we present an investigation of the stellar content of ngc3576 through the @xmath16 , @xmath17 and @xmath18 imaging and @xmath1band spectroscopy ( described in 2 ) . in
3 we discuss our results , and our conclusions are summarized in 4 .
@xmath16 ( @xmath19 @xmath20 1.3 @xmath2 m , @xmath21@xmath19 @xmath20 0.3 @xmath2 m ) , @xmath17 ( @xmath19 @xmath20 1.6 @xmath2 m , @xmath21@xmath19 @xmath20 0.3 @xmath2 m ) and @xmath18 ( @xmath19 @xmath20 2.1 @xmath2 m , @xmath21@xmath19 @xmath20 0.4 @xmath2 m ) images of ngc3576 were obtained on the nights of 1999 march 3 and 4 and 2000 may 19 and 20 with the f/14 tip tilt system on the cerro tololo interamerican observatory ( ctio ) 4-m blanco telescope using the facility imager osiris ) and on the nights of 1998 july 9 to 13 with the facility imager cirim ) mounted on the 1.5-m telescope .
spectroscopic data were obtained with the blanco telescope and the facility near infrared spectrometer , irs , in 1998 may 17 and june 23 , and with osiris using the f/14 tip - tilt system in 1999 march 34 , 1999 may 23 , and 2001 july 7 and 12 .
osiris delivers a plate scale of 0.16@xmath22/pixel , the irs 0.32@xmath22/pixel , and cirim 1.16@xmath22/pixel .
all basic data reduction was accomplished using iraf .
each image was flat - fielded using dome flats and then sky subtracted using a median - combined image of five to six frames .
independent sky frames were obtained 510@xmath23 south of the ngc3576 cluster .
the osiris 1999 march images were obtained under photometric conditions .
total exposure times were 180s , 45s and 45s at @xmath16 , @xmath17 and @xmath18 , respectively .
the individual @xmath16 , @xmath17 and @xmath18 frames were shifted and combined .
these combined frames have point sources with fwhm of @xmath20 0.61@xmath24 , 0.88@xmath24 and 0.64@xmath24 at @xmath16 , @xmath17 and @xmath18 , respectively .
dophot @xcite photometry was performed on the combined images .
the flux calibration was accomplished using standard star gspc s427-d ( also known as [ pmk98 ] 9123 ) from @xcite which is on the las campanas observatory photometric system ( lco ) .
the lco standards are essentially on the cit / ctio photometric system @xcite , though color transformations exist between the two systems for redder stars .
no transformation exists between osiris and either cit / ctio or lco systems .
the standard observations were made just after the ngc3576 data and within 0.24 airmass from the target .
no corrections were applied for these small difference in airmass .
aperture corrections measured inside 20 pixel radius circles were used to put the instrumental magnitudes on a flux scale .
six uncrowded stars on the ngc3576 images were used for this purpose .
since the brightest stars in the 1999 images were saturated , we have taken short exposure images in may 2000 . although the conditions in may 2000 were non - photometric , we used stars in common with the 1999 images to determine the zero point for the additional ( bright ) stars .
uncertainties for the @xmath16 , @xmath17 and @xmath18 magnitudes in 1999 images include the formal dophot error added in quadrature to the error in the mean of the photometric standard and to the uncertainty of the aperture correction used in transforming from the dophot photometry to osiris magnitudes .
the sum in quadrature of the aperture correction and standard star uncertainties are @xmath250.032 , @xmath250.034 and @xmath250.069 mag in @xmath16 , @xmath17 and @xmath18 , respectively .
the scatter in the instrumental magnitudes in the set of stars from the 1999 images used to calibrate the may 2000 images are @xmath250.010 ( @xmath16 ) , @xmath250.012 ( @xmath17 ) and @xmath250.010 ( @xmath18 ) mag ; thus the errors in the bright star magnitudes are dominated by the uncertainty in the standard stars .
the dophot errors were larger than @xmath25 0.01 mag , and we adopted a cutoff for errors larger than 0.05 mag , which corresponds to a limiting magnitude of @xmath26 @xmath27 15.0 .
lower angular resolution images were obtained at @xmath16 , @xmath17 and @xmath18 using cirim at f/8 on the ctio 1.5-m telescope(1.16@xmath24 pixel@xmath10 ) .
the individual frames in each filter were shifted and combined and have measured seeing of 2.2@xmath24 fwhm .
although collected under photometric conditions , these images are not as deep as that of the 4-m telescope and were used only to transform to equatorial coordinates , since they encompassed a wider field than the osiris images taken at the 4-m telescope .
the @xmath1band spectra of eight of brightest stars in the ngc3576 cluster were obtained : # 11 , # 48 , # 69 and # 160 with the irs and # 4 , # 48 , # 78 , # 95 and # 184 with osiris .
the spectra were divided by the average continuum of sevaral a type stars to remove telluric absorption .
the br@xmath28 photospheric feature was removed from the average a type star spectrum by eye by drawing a line between two continuum points .
one dimensional spectra were obtained by extracting and summing the flux in @xmath25 2 pixel aperture .
the extractions include background subtraction from apertures , 1 - 2@xmath22 on either side of the object .
the wavelength calibration was accomplished by measuring the position of bright oh@xmath29 lines from the @xmath1band sky spectrum @xcite .
the spectral resolution at 2.2 @xmath2 m is @xmath19/@xmath21@xmath19 @xmath20 3000 for osiris and @xmath19/@xmath21@xmath19 @xmath20 825 for the irs .
the osiris @xmath16 , @xmath17 and @xmath1band images reveal an embedded star cluster .
we detected 315 stars in the @xmath1band to a limiting magnitude of 16 ( see below ) in a region of 0.03 square degrees .
figure [ finding ] shows a finding chart using the @xmath1band image .
a false color image is presented in figure [ color ] , made by combining the three near infrared images and adopting the colors blue , green and red , for @xmath16 , @xmath17 , and @xmath18 , respectively . in this way , the bluest stars are likely foreground objects , and the reddest stars are probably @xmath1band excess objects , indicating the presence of hot dust for objects recently formed in the cluster ( background objects seen through a high column of interstellar dust would also appear red ) .
the diffuse nebula is mainly due to br@xmath28 emission in the h ii region .
the dark patches in the bottom right of figure [ color ] are zones of the giant molecular cloud from which ngc3576 is emerging .
there is no doubt that this is a signature of a young cluster containing massive stars , now in the process of shredding the local molecular cloud .
the cluster is asymmetric , with the majority of the stars encompassed in a semi circle ; and there is a definite appearance that the cluster is destroying the cloud from the ne to the sw .
the ne is also the direction toward which the h ii region is seen in the visual . the @xmath30 _ versus _
@xmath18 color magnitude diagram ( cmd ) is displayed in figure [ cmd ] .
the open circles indicate objects fainter than @xmath31 .
a concentration of dots appear around @xmath32 , probably indicating the average color of cluster members .
a number of stars display much redder colors , especially the brightest ones .
the solid vertical line indicates the theoretical main sequence ( see below ) . the @xmath33 _ versus _
@xmath30 color color plot is displayed in figure [ ccd ] .
open circles indicate objects fainter than @xmath34 .
open triangles indicate stars fainter than @xmath31 and @xmath34 .
the numbers labeling stars in both plots refer to the same objects .
the inclined lines , from top to bottom , indicate interstellar reddening directions for main sequence m
type @xcite , o
type @xcite and t tauri @xcite stars .
stars to the right of the solid line deviate from pure interstellar reddening , probably because of hot dust emission .
the effect of this excess emission is stronger in the @xmath1band than at shorter wavelengths .
although the open circles and triangles in the bottom right of figure [ ccd ] indicate only lower limits , these objects are also likely affected by thermal emission .
they are bright in the @xmath1band and should be detected in @xmath16 and/or @xmath17 if affected only by interstellar reddening and if they were cluster members with typical extinction .
the cluster characteristics indicate that the stars havent had enough time to evolve away from the main sequence .
we expect that most of the high mass stars are close to the zero age main sequence ( zams ) .
we can estimate the reddening toward the cluster from a simple approximation @xcite @xmath35 @xmath27 1.6@xmath36 and using the fact that the average intrinsic color of hot stars is almost zero @xcite .
the stars brighter than @xmath37 have an average color of @xmath38 , corresponding to @xmath39 = 1.57 mag ( @xmath40 @xmath20 15.7 mag ) .
the interstellar component of the reddening can be separated from that local to the cluster stars by using the star hd 97499 .
this is a foreground star , since it is brighter and less reddened than stars in the cluster and is offset from the radio source line - of - sight by 2 arc - minutes . from its spectral type b1 - 2iv - v _ michigan spectral catalogue _ - @xcite and magnitude , a distance of 2.4 kpc was derived @xcite .
our measurements of this star result in @xmath41 , indicating a@xmath42 .
this gives a@xmath43mag kpc@xmath10 , which is close to the expected extinction for this position along the galactic plane . for 2.8 kpc distance of ngc3576 , the interstellar component is then a@xmath44 , leaving a local component of a@xmath45 .
we were not able to independently check the distance derived by depree ( 1999 ) via spectroscopic parallax ( see below ) , but we expect that the radio kinematic distance is reliable for this galactic direction . in order to place the zams in the cmd , the corresponding bolometric magnitudes ( @xmath46 ) and effective temperatures ( @xmath47 ) must be transformed into @xmath30 colors and apparent @xmath18 magnitudes .
@xcite models for the zams with solar abundances were adopted .
the bolometric corrections ( bc ) applied to derive absolute visual magnitudes ( @xmath48 ) are from @xcite for spectral types @xmath49 to @xmath50 and from @xcite for later spectral types .
the intrinsic colors ( @xmath51 , @xmath52 , and @xmath53 ) are from @xcite . using the distance modulus ( 12.24 ) and the apparent visual magnitude
, we transform the @xmath46 into apparent @xmath54 magnitudes .
the correspondence between spectral types and @xmath47 are from @xcite for @xmath49 to @xmath50 and from @xcite for later spectral types .
since koornneef s colors are in the johnson system , which is nearly identical to saao system @xcite , we used the saao to cit / ctio relations to transform koornneef s @xmath30 color indices to the cit / ctio system .
these corrections are about 1% and could be neglected when compared with the photometric errors and differential reddening .
the zams is represented by a vertical solid line in figure [ cmd ] , shifted to @xmath55 kpc and reddened by a@xmath56 = 0.43 due to the interstellar component . when adding the average local reddening ( a@xmath57 , the zams line is displaced to the right and down , as indicated by the dashed lines .
we can not fix the position of the zams , since there is a scatter in the reddening .
the small group of relatively bright stars ( @xmath58 ) in between these two lines , suggests that some of them , the bluer ones , could mark the position of the zams .
unfortunately , we do nt yet have spectra of these objects to check whether or not this is true .
objects to the right of the o - type stars line ( figure [ ccd ] ) have colors deviating from pure interstellar reddening .
this is frequently seen in young star clusters and is explained by hot dust in the circumstellar environment .
we can estimate a lower limit to the excess emission in the @xmath1band by supposing that the excess at @xmath16 and @xmath17 are negligible , and that the intrinsic colors of the embedded stars are that of ob stars .
indeed , assuming that our sample of stars is composed by young objects ( not contaminated by foreground or background stars ) , any object would have an intrinsic color in the range @xmath59 mag @xcite .
let us adopt for all objects in our sample the intrinsic colors of a b2 v star : @xmath60 and @xmath61 @xcite .
the error in the color index would be smaller than the uncertainty in the mathis law , we are using for the interstellar extinction .
> from the difference between the observed @xmath33 and the adopted b 2 v star we obtain the color excess and by using the relation ( @xmath62 ) we derive the interstellar extinction in the @xmath63band and so the intrinsic apparent magnitudes @xmath64 , @xmath65 and @xmath54 .
> > from another relation of mathis law ( @xmath66 ) , we derive the magnitude at @xmath1band corrected from the interstellar extinction .
the difference between this number and @xmath54 gives the excess emission in the @xmath1band , due to dust thermal emission . for stars not detected in the @xmath63band
we suppose they are affected by an interstellar extinction @xmath67 equal to the median value of those measured in the @xmath63band . in this way , we derive the extinction in the @xmath68band by using mathis law in the form : @xmath69 . from here the procedure to derive the @xmath1excess magnitude
follows the same steps as we did before . for stars not detected in the @xmath68band
, we still want to estimate the @xmath1excess , since potentially interesting objects are too red to be detected in the @xmath68band .
a lower limit for this excess can be derived by using the same procedure as above , but assigning a limiting magnitude @xmath70 for objects not detected in the @xmath68band .
our results are displayed in figure [ kexc ] .
objects with very large excess in the upper right corner of figure [ kexc ] , can not be explained by errors in the de reddening procedure and could be real .
they could represent the emission of accreting disks around the less massive objects of the cluster . in order to separate the cluster members from projected stars in the cluster direction , we imaged a region close to ngc3576 .
the star counts were normalized by the relative areas projected on the sky and then binned in intervals of @xmath71 and @xmath72 .
the stellar density in the field was then subtracted from that of the cluster in bins of magnitude and color intervals .
this works well for foreground stars , since there were a few in the cluster field .
regarding the background , the situation is more complex , however , since ngc3576 produces so high an obscuration , almost no background objects can be seen at this limiting magnitude .
the completeness of dophot detections was determined through artificial star experiments .
this was accomplished by inserting fake stars in random positions of the original frame , and then checking how many times dophot retrieved them .
the psf of the fake star was determined from an average of real stars found in isolation and in areas of dark sky . in total , we inserted 2400 stars in the magnitude interval 8 @xmath73 k @xmath73 20 , which amounts to six times the number of real stars recovered in the original dophot run .
for every @xmath74 we inserted simultaneously 5 stars , repeating the procedure for 20 times .
the insertion of all the 100 stars at once would impact the stellar crowding and change the detection conditions .
the incompleteness of the sample is defined as the percentage of times the fake star fails to be recovered .
we performed these experiments in the whole frame and also in each of the three sub images that were cut out from it , displayed in figure [ finding ] . the upper right sub
image is representative of detection limited by photon statistics , the one at the center , by high background and the lower left one , by stellar crowding . in figure [ completeness ]
we present the photometric completeness .
the _ dashed _ line ( without symbols ) refers to the whole image .
the limit is different for different sub images , e.g. for an area with high nebular background ( _ circles _ ) , for a crowded area ( _ squares _ ) , or for an area with few stars and a dark sky ( _ triangles _ ) . the performance of the photometry is better than 90% for a 15th magnitude star found in isolation , as compared to stars in the nebular zone which need to be ten times brighter to be detected with the same efficiency
since there is no objective way to define the sub - image limits we applied a single completeness correction to the whole frame ( dashed line ) . as seen in figure [ completeness
] such a correction is close to the curve limited by crowding .
future work seeking to obtain deeper photometry in this cluster demands a substantial improvement of the spatial resolution and will require adaptative optics imaging from ground based telescopes .
after correcting for non - cluster members , interstellar reddening , excess emission ( a lower limit ) and photometric completeness , the resulting @xmath1band luminosity function ( @xmath75 ) is presented in figure [ klf ] .
a linear fit , excluding deviant measures by more than @xmath76 , has a slope @xmath77 .
a similar klf slope was obtained for w42 ( @xmath78 ) by @xcite .
we can evaluate the stellar masses by using @xcite models , assuming that the stars are on the zams and not the pre - main sequence ( but see below the discussion in 3.2 ) .
this is a reasonable approximation for massive members of such a young cluster .
the main errors in the stellar masses are due to the effects of circumstellar emission and stellar multiplicity .
our correction to the excess emission is only a lower limit , since we assumed the excess was primarily in the @xmath18 band . @xcite
have computed disk reprocessing models which show the excess in @xmath16 and @xmath17 can also be large for disks which reprocess the central star radiation .
in general , we can expect the excess emission to result in an overestimate of the mass of any given star and the cluster as a whole .
the slope of the mass function should be less effected .
it is difficult to quantify the effect of binarity on the imf .
if a given source is binary , for example , its combined mass would be larger than inferred from the luminosity of a `` single '' star and its combined ionizing flux would be smaller .
the cluster total mass would be underestimated , the number of massive stars and the ionizing flux would be overestimated .
the derived imf slope would be flatter than the actual one . with these limitations in mind
we have transformed the klf into an imf . since other authors also do not typically correct for multiplicity , our results can be inter compared , as long as this parameter does nt change from cluster to cluster .
the @xmath79 slope derived for ngc3576 is @xmath4 ( figure [ imf ] ) , which is consistent with salpeter s slope @xcite . a similar imf slope was obtained for the trapezium cluster ( @xmath80 ) by @xcite .
flatter slopes have been reported only for a few clusters , most notably the arches and quintuplet clusters @xcite , both near the galactic center .
flatter slopes may indicate that in the inner galaxy star forming regions , the relative number of high mass to the low mass stars is higher than elsewhere in galaxy .
it is also possible that dynamical effects may be more important in the inner galaxy . @xcite
have modeled the arches cluster data with a normal imf , but include the effects of dynamical evolution in the presence of the galactic center gravitational potential .
they find the observed counts are consistent with an intial salpeter like imf .
we derived an upper limit to the total mass of the cluster ( our imf is likely overestimated due to excess emission , see above ) by integrating the @xmath79 between @xmath81 - where the distribution is nearly continuous .
the lower mass limit was adopted from @xcite taking into account the @xmath79 turnover measured by those authors in orion .
the integrated cluster mass is m@xmath82 = 5.4 x 10@xmath83 m@xmath84 .
as pointed out above , this is likely an upper limit .
source # 48 is anomalously bright and needs to be treated separately .
let us assume that it is affected by an interstellar reddening equal to the cluster average : a@xmath85 , which implies a derredned magnitude k@xmath86 .
if we take at face value the @xmath87 color to represent only an excess at @xmath18 , the reprocessing disk would contribute with @xmath88 .
this excess emission is clearly an underestimation , since the stellar flux is swamped by the disk emission to the point of veiling all the photospheric lines ( see below ) .
moreover , if the excess emission was so small , the luminosity of # 48 would require a cluster of four 100 m@xmath84 stars , unresolved down to a limit of 0.6@xmath22 , which does nt seem to be the case .
if object # 48 is a single o3 v star , it would contribute with nlyc @xmath12 1.17 @xmath25 0.05 x 10@xmath9 s@xmath10 .
alternately , we can evaluate the excess emission by using @xcite models for reprocessing disks . by starting with the maximum excess emission @xmath89 valid for a o7-type star ( their table 4 ) we derive the m@xmath90 .
using vacca et al .
( 1996 ) calibration , we obtain the corresponding stellar spectral type ( and mass ) that is much smaller than the o7-type we started with .
the next step is reducing the excess emission ( adequate for a smaller stellar luminosity ) , deriving a larger final mass , and iterating until convergence .
this was achieved for an excess emission @xmath91 , corresponding to a spectral type b1 v and mass m @xmath12 17 m@xmath84 .
this is possibly a lower limit , since the intervening extinction toward # 48 is probably larger than the cluster average .
we can say that # 48 is a late o / early b / early b yso , very similar to what has been found by @xcite in w31 for the brightest @xmath1band object in that cluster .
object # 48 is buried in a large and dense disk , and its contribution to the cluster mass and ionizing photons are negligible . we have compared the locations of the @xmath1band sources in our images with the mid infrared sources of @xcite .
source # 48 is very close to their irs1 and # 11 to their irs3 .
these sources appear also in the iras small scale structure catalogue ( x1109 - 610 ) and the iras point source catalogue ( iras11097 - 6102 ) . in the case of irs1 , at least , recent high resolution mid - infrared images @xcite clearly indicate that it is associated with our object # 50 , 1@xmath22 to the south of the @xmath1band source # 48 . the number of lyman continuum photons derived from the imf , excluding object # 48 is nlyc @xmath12 0.42 @xmath25 0.22 x 10@xmath9 s@xmath10
. the contribution of this single object could be as large as nlyc @xmath12 1.17 x 10@xmath9 s@xmath10 , in the case it is an o3 v star .
this is very close to the nlyc @xmath12 1.6 @xmath25 0.4 x 10@xmath9 s@xmath10 derived from radio observations @xcite scaled to a distance of 8 kpc to the galactic center .
however , we have shown in the preceeding sub - section , object # 48 probably is a much less massive object , a b1 v star .
it can be seen that the nlyc is highly sensitive to the particular procedure used to correct for the excess emission and extinction .
very probably we have missed a handful of main sequence o - type stars responsible for the ionizing flux seen at radio wavelengths . moreover , the spectra of the eight brightest stars described in the next section , indicate that the ionizing stars must be apparently faint .
the properties of the cluster are summarized in table 1 .
the stellar cluster is located in the ne border of a molecular cloud .
the stellar density increases toward the sw , ending abruptly , with a few sources embedded in the molecular cloud ( figure [ color ] ) .
the spatial distribution of color indices ( figure [ cc_map ] ) also shows a similar gradient . this is due , in part , to the increasing extinction toward the inner molecular cloud .
however , the red colors are intrinsic to many of the sources , since there are excess emission objects in this zone .
this suggests that stars at the sw are younger than at ne , since recent models by @xcite predict formation times of the same order for stars of different masses .
a similar scenario for this cluster , by which star formation is progressing toward the inner zones of the molecular cloud , has been suggested by @xcite and @xcite and our images dramatically confirm this to be the case ; see figure [ color ] .
spectra of eight cluster members ( # 4 , # 11 , # 48 , # 69 , # 78 , # 95 , # 160 and # 184 ) , are shown in figure [ co ] and figure [ ftless ] .
spectra of a foreground m - type and an a - type stars were added at the top of the figures for comparison .
the a - type star also was divided by the average continuum slope of the other observed a - type stars .
object labels are the same as in figure [ finding ] ; [ cmd ] ; and [ ccd ] .
ordinates in figure [ co ] and figure [ ftless ] are normalized fluxes , as follows .
we constructed templates for telluric absorption bands by observing a - type stars , close in time and airmass to the target stars .
we removed the br@xmath28 line from those spectra by hand ( linear interpolation ) , since this region is free from telluric features .
there are no other noticeable photospheric features in a - type stars in the @xmath1band .
then we divided the spectrum of each target by the appropriate a - type spectrum and normalized the resulting spectrum at 2.19 @xmath2 m .
the signal to noise ratio is s / n @xmath92 for # 4 , # 11 and # 95 and @xmath93 for the other objects . the spectra were placed on a flux scale by dividing by @xmath94 and multiplying by @xmath1band fluxes corresponding to the @xmath1magnitudes in table 2 .
however , it is straight forward to compare the observered spectra ratioed only by the a type continuum .
it can be seen that all the cluster member candidates display rising continuua to the red when compared to the a type and m type foreground stars .
these eight stars are thus most likely cluster members , but we can not rule out that some may be background stars
. we must take into account that these objects are projected toward the central part of the cluster and thus subject to high obscuration ( a@xmath85 ) due to the intra cluster gas and dust .
objects # 48 , # 95 , and # 160 are undoubtedly cluster members , since , in addition , they have excess emission .
object # 69 is bright in the @xmath17 and @xmath18 bands and judging from its @xmath30 color , it should be relatively bright in the @xmath63band , but it is nt detected at all in the @xmath63band osiris images , appearing above the reddening line in figure [ ccd ] .
the spectrum of this object in figure [ ftless ] looks like its neighbor # 78 , that is 1.5 magnitude fainter and still is detected in the @xmath63band .
this puzzling situation was clarified when we took a @xmath1band acquisition image ( feb/2002 ) with phoenix at gemini under very good seeing ( @xmath27 0.250.3 ) . in that image ,
object # 69 is shown as a small nebula , with no sign of a buried point - like source .
object # 69 is a clump of dense material ionized by object # 48 or some other neighboring source
. the h@xmath95 2.122 @xmath2 m emission could be either shock or ionization produced .
the narrow features at 2.058 ( ) and 2.166 @xmath2 m ( br@xmath28 ) in figures [ co ] and [ ftless ] are due to contamination from the extended nebula .
the spectra were extracted in such a way that the large scale nebular component is over subtracted . in some objects these lines are in emission , due to enhanced nebular emission close to the star .
objects # 48 , # 160 , # 78 and # 69 show h@xmath95 2.122 @xmath2 m in emission .
this feature appears is emission also in # 11 and # 95 , but since the he i and br@xmath28 components are not over subtracted , this feature may be due to contamination of extended nebular emission .
the co bandhead at 2.2935 @xmath2 m is in absorption in # 4 , # 160 , # 184 and in emission in # 48 .
none of those objects show photospheric lines indicating that they are still enshrouded in their birth cocoons .
this is corroborated by the excess emission in the @xmath1band derived from photometry except for objects # 4 and # 184 ( table 2 ) .
a variety of mechanisms and models have been proposed to explain the origin of co emission in ysos .
these include circumstellar disks , stellar or disk winds , magnetic accretion mechanisms such as funnel flows , and inner disk instabilities similar to those which have been observed in fu orionis like objects and t tauri stars in a phase of disk accretion @xcite .
@xcite shows that gas free - falling along the field lines yield the bandhead profiles , in agreement with those observed , with the shape of the profile determined mainly by inclination of the disk to the line of sight .
hanson et al . ( 1997 ) reported the presence of co in emission in several masssive stars in m17 .
the situation is less clear for objects # 4 , # 160 and # 184 displaying co in absorption .
the absence of large color excess indicates that they could be cool pre - main sequence stars still in a contraction phase . while we can not rule out that they are evolved background m - type stars , they do appear projected on the core of the newly formed cluster .
these objects deserve further study , and if they are indeed pre - mainsequence stars , then the imf determination above has a component whose masses have been overestimated .
the h@xmath95 molecular emission is produced by shocks and may indicate the existence of gas outflow . since we subtracted the extended background close to the stars , the spectra show only the spatially unresolved component of h@xmath95 .
we can thus be confident that h@xmath95 emission in # 48 and # 69 are emitted close to the stellar sources .
however , since these two sources are only @xmath96 apart , the emission could be associated with either or both objects .
it may be surprising that most of the stars with featureless spectra are close to the interstellar reddening line .
however this is in accord with the hillenbrand et al .
models ( 1992 ) for early type stars , which predict relatively small color excesses ( @xmath97h - k@xmath98 ) for objects with large excess emission ( @xmath99 ) .
such models do not predict large departure from the reddening line , like displayed by objects # 160 and # 50 . those sources might be surrounded by local dusty clouds , in addition to the accreting disk .
we have presented deep @xmath16 , @xmath17 and @xmath18 images of the newborn stellar cluster in ngc3576 ( figure [ color ] ) and @xmath1band spectra for eight cluster members . the @xmath1band excess emission displayed by objects # 4 , # 11 ( irs3 ? ) , # 48 , # 69 , # 78 ,
# 95 , # 160 and # 184 , in combination with their featureless continuum or co emission / absorption ( figure [ co ] ) , indicates that they are young , massive stars still in the process of accreting material from their birth cocoons .
the lack of photospheric features and presence of disk signatures indicates that ngc3576 is one of youngest massive star clusters in the milky way .
our data also confirm the scenario of star formation progressing from the ne toward the inner parts of the molecular cloud ( sw ) , but the very young age of the cluster may contradict the claim of enhanced he abundance in the ne part of the gh ii region by @xcite .
there are no evolved stars in the cluster which could produce the enhancement and the time to diffuse nuclear processed material from neighboring regions into the nebular environment would be much larger than the cluster age .
the fact that the nebular excitation increases toward the ne , producing stronger he lines , may have influenced the abundance calculations .
since our data do not enable us to derive spectroscopic parallaxes ( no photospheric lines were detected in the luminous stars ) , we have adopted the radio distance obtained by @xcite revised to 2.8 kpc by using the most recent galactic center distance ( @xmath8 kpc , @xcite ) .
the cluster parameters are not well constrained because of distance uncertainty and the difficulty in correcting the @xmath1band magnitude for the circumstellar dust emission and extinction .
the overall picture summarized in table 1 of a massive and dense cluster remains valid .
however , the ionizing flux derived from the imf is much smaller than that from radio observations .
this is consistent with the fact that we havent found spectroscopically the massive main sequence stars that ionize the cluster .
those stars probably remain behind heavily obscuring clouds . the fact that several of the brightest cluster members do not show revealed photospheres raises the question : where are the ionizing sources of ngc3576 ?
it is plausible that some of the faint sources are in reality luminous objects seen through large extinction , and were not accurately dereddened because they escaped detection in the @xmath16 and @xmath68band .
this may also explain why the number of lyman continuum photons derived from the imf is smaller than that measured at radio wavelengths .
to tackle this question , we plan to obtain spectra from stars of @xmath100 and fainter using the gemini south 8m telescope . in order to understand better the circumstellar environment of the ysos we examined here , we have performed high resolution mid - infrared imaging with the gemini south telescope which will be analyzed in a future paper .
our goal is to derive the characteristics of the circumstellar dust , in particular evaluating its contribution to the near infrared excess emission , extinction , and the possible presence of accreting disks .
ll cluster distance & 2.8 @xmath25 0.3 kpc + cluster diameter & 1.5 pc + cluster mass & @xmath1015.4 x 10@xmath83 m@xmath84 + stellar density & @xmath1013.1 x 10@xmath83 m@xmath84 pc@xmath102 + imf slope & @xmath103 + nlyc phot .
( imf ) & 0.42 - 1.67 x 10@xmath9s@xmath10 + nlyc phot .
( radio ) & 1.6 @xmath25 0.4 x 10@xmath9s@xmath10 + lrcrrl # 4 & 1.51 & 0.93 & 9.95 & -0.01 & co abs + # 11 & @xmath1040.99 & 1.10 & 12.91 & -0.26 & + # 48 & 3.04 & 2.21 & 8.35 & -3.17 & h@xmath95 & co em + # 69 & @xmath1043.44 & 1.45&10.10 & -0.62 & h@xmath95 em + # 78 & 2.02 & 1.38&11.67 & -0.16 & h@xmath95 em + # 95 & 1.67 & 1.41 & 9.26 & -0.40 & + # 160 & @xmath1040.95 & 2.93 & 11.12 & -2.09 & co abs + # 184 & 1.23 & 0.81&10.40 & -0.07 & co abs + | we present deep , high angular resolution near infrared images of the obscured galactic giant h ii region ngc3576 .
our images reach objects to @xmath0 .
we collected high signal to noise @xmath1band spectra of eight of the brightest objects , some of which are affected by excess emission and some which follow a normal interstellar reddening law .
none of them displayed photospheric features typical of massive ob type stars .
this indicates that they are still enshrouded in their natal cocoons .
the @xmath1band brightest source ( ngc3576#48 ) shows co 2.3 @xmath2 m bandhead emission , and three others have the same co feature in absorption .
three sources display spatially unresolved @xmath3 emission , suggesting dense shocked regions close to the stars .
we conclude that the remarkable object ngc3576#48 is an early b / late o star surrounded by a thick circumstellar disk / envelope .
a number of other relatively bright cluster members also display excess emission in the @xmath1band , indicative of reprocessing disks around massive stars ( ysos ) .
such emission appears common in other galactic giant h ii regions we have surveyed .
the imf slope of the cluster , @xmath4 , is consistent with salpeter s distribution and similar to what has been observed in the magellanic cloud clusters and in the periphery of our galaxy . |
as stars evolve , the energies from stellar winds and supernova ( sn ) explosions ( hereafter collectively referred to as `` feedback '' ) are released into the surrounding interstellar material .
the physics of this interaction is complex , with dependencies on ( at least ) the local neutral and molecular gas density and porosity , the local and total gravitational potential depth and shape , the intensity and duration of massive star formation ( sf ) , and the metal content of the parent cloud . in a simplistic scenario , a single massive star forms and evolves , unleashing a characteristic energy of @xmath010@xmath2 erg per sn event @xcite .
a fraction of this energy is converted into bulk motion of the surrounding material .
this energy can be compared with the amount of energy required to move that same gas mass a certain distance within a given gravitational potential .
if the feedback energy exceeds the requisite formation energy , then stellar evolution is capable of creating structures in the interstellar medium ( ism ) . scaling this simple single - star scenario
up to bursts of sf , the conditions for the creation of macroscopic ism features could be idealized in the case of an intense and rapid sf event @xcite .
this would result in the simultaneous and coherent injection of feedback energy from multiple stars .
one would expect to find coherent ism structures , such as h i holes and shells , with remnant stellar clusters within ; those cluster ages would in principle be similar to the age of the ism feature that surrounds it .
recent results have challenged this simplistic model . in systems littered with coherent h i structures ( e.g. , holmbergii ; see discussion in weisz etal .
2009b ) , there are few instances of high surface brightness remnant clusters ( see also rhode etal .
however , deep _ hubble space telescope _
( _ hst _ ) imaging reveals young stellar populations within all h i features ; energy estimates based on spatially resolved star formation history ( sfh ) analyses show that the energy requirements for the creation of those ism features is easily met by stellar winds and supernovae .
importantly , this energy is derived from sf spanning hundreds of myr and encompassing multiple episodes of sf within a given region . a new scenario supported by recent
work is one where the present - day morphology and dynamics of a given galaxy are the results of the cumulative sf over at least the past few hundred myr .
the properties of the stellar populations in ddo165 make it a particularly important test case for this model of the sf - ism interaction .
as the color _ hst _ image in figure [ figcap1 ] shows , ddo165 contains rich populations of luminous stars , both blue and red . the global color magnitude diagram ( cmd ; see figure [ figcap2 ] ) derived from _
data shows that the system has been in a burst phase ( defined as the sf rate at recent times exceeding the average past rate by more than factor of two ) for more than 1 gyr ( see details in * ? ? ?
this easily exceeds the dynamical timescale ( 710@xmath340 myr ) and is among the longest starburst episodes known to date in any dwarf galaxy .
the recent ( up to 500 myr ) sfh shows significant variations on timescales @xmath450 myr , including a precipitous drop in sf activity within the past 25 myr ( see detailed discussion below ) .
further , @xcite find that ddo165 has a long characteristic timescale for structure evolution in its stellar populations ( at least 350 myr ; see further discussion below ) ; once formed , stellar associations remain coherent for a large fraction of a dynamical timescale . the neutral ism dynamics of ddo165 ( see cannon etal . 2011
; hereafter paper i ) are equally intriguing in the context of the feedback model .
the h i morphology is dominated by a giant h i hole ( diameter @[email protected] kpc ) that is slightly offset from the center of the stellar distribution .
seven other h i holes are cataloged in paper i ; some of these have measurable expansion velocities , while others are static at the present time .
the h i gas is concentrated in two kinematically distinct features : a southern , main component and a northern component that contains @xmath015% of the total h i mass .
based on the h i data alone , we can exclude many potential origins for this kinematic discontinuity ( e.g. , coherent solid - body rotation ; superposed or counter - rotating disks ; turbulence ) , but both infall ( i.e. , interaction ) and outflow ( i.e. , `` blowout '' ) models remain viable .
the southern h i component harbors multiple localized high velocity features , each of which spans up to @xmath060 kms@xmath5 in velocity space but is no larger than the h i beam width ( @xmath0430 pc ) . taken together , these peculiar properties of ddo165 make it a promising laboratory for examining the effects of feedback on the surrounding interstellar material . to that end , the present work investigates the stellar populations throughout ddo165 , exploiting the spatially resolved _ hst _ data to study those populations associated with each of the features identified in the h i analysis of paper i. in a broad sense , we seek to understand if spatially and temporally extended sf is capable of creating giant h i holes , high velocity gas features , and kinematically distinct gas components .
ddo165 was observed with the _ very large array _ ( _ vla _ ) in the h i spectral line during three observing sessions between november , 2006 and november , 2007 for program ac842
the treatment of these data , including reductions , handling and analyses , are described fully in paper i. ddo165 was observed with the _ hst _ on march 17 , 2006 , as part of observing program go-10605 ( p.i .
skillman ) . the treatment and reduction of these images are discussed in detail in @xcite . over 144,000 stars
are measured using the dolphot package @xcite ; the 50% completeness levels are [email protected] ( [email protected] ) and [email protected] ( [email protected] ) .
figure [ figcap1 ] shows a color representation of the _ hst_/acs field of view of ddo165 .
all of the high surface brightness stellar populations are covered by this single pointing .
note the pronounced truncation of the young stellar population in the southern region of the system ( see also further discussion in [ s4.1 ] ) .
the brightest objects in the field are galactic foreground stars ; other compact high surface brightness objects are stellar clusters within ddo165 ( see further discussion in [ s3.3.3 ] below ) or background galaxies .
the v vs. ( v@xmath10i ) cmd of all 144,234 stars within the _ hst_/acs field of view is shown in figure [ figcap2 ] .
five prominent stellar evolution phases are indicated : the main sequence ( ms ) , the blue helium burning sequence ( bheb ) , the red helium burning sequence ( rheb ) , the asymptotic giant branch ( agb ) , and the red giant branch ( rgb ) .
note the strengths of the bheb and rheb sequences compared to the ms .
the bheb is @xmath02 magnitudes brighter than the ms at a given stellar age ; thus , the prominence of the helium burning sequences suggests stronger sf in the recent past than at the current time .
in this work , we focus on the bheb stars because of their unique relationship between age and absolute magnitude . as discussed in detail in @xcite and multiple works since then , the ages of bheb stars increase as the absolute magnitude increases .
further , the age gradient becomes shallower for increasingly faint bheb stars , resulting in degraded temporal resolution at larger look back times ; for example , 10 myr resolution is available back to @xmath0100 myr , but @xmath050 myr time bins are required when going back to 500 myr .
the bheb merges with the red clump at ages @xmath11600 gyr ; at larger lookback times , other regions of the cmd must be used . in the present work we concentrate on the most recent 500 myr ,
during which the sfh is quite robust .
the observed cmds are discussed here and in [ s3.2 ] ; these cmds are modeled using maximum likelihood techniques to extract the intensity of sf over the last 500 myr in [ s3.3 ] .
we begin to explore the rich information in the global cmds by extracting the ages of bheb stars in 5 coarse age bins : 0 - 25 myr ( though note that the bheb sequence is insensitive to ages @xmath410 myr ) , 25 - 50 myr , 50 - 100 myr , 100 - 200 myr , and 200 - 400 myr . since the positions of these stars are known , we can then plot the location of bheb stars in each age bin to allow a simple snapshot of the patterns of sf within ddo165 over the last few hundred myr .
figure [ figcap3 ] shows this plot ; the positions of bheb stars of various ages are color - coded for clarity . overlaid in black on each panel is the h i column density contour at the 10@xmath12 @xmath13 level , as well as the locations of the six h i holes identified in paper i that fall completely within the _ hst_/acs field of view ( see more detailed discussion below ) .
two important general features of the sf patterns in ddo165 are apparent from figure [ figcap3 ] .
first , the total number of bheb stars rapidly increases when considering successively older age bins .
276 bheb stars are measured in the 25 - 50 myr time frame ; this increases to @xmath141,100 stars with ages between 50 and 100 myr .
this is partially due to imf effects , but is also suggestive of a substantial sf event during the past 100 myr , with more stars forming per unit area around 100 myr ago than at the present time .
second , in all epochs older than @xmath0100 myr , sf pervades the entire disk of ddo165 .
this supports the spatially and temporally extended burst scenario described above .
figure [ figcap3 ] demonstrates that ddo165 has a complex history of sf even over the most recent 500 myr . from figure [ figcap2 ] ,
the tip of the red giant branch ( trgb ) occurs at m@xmath15 @xmath16 [email protected] .
assuming m@xmath15 = @[email protected] at the trgb @xcite and e(b@xmath10v ) @xmath16 0.024 mag @xcite , the distance of ddo165 derived from our _ hst _ data is in good agreement with the value found by @xcite and adopted in paper i , d @xmath16 [email protected] mpc .
we thus retain their distance estimate throughout the present work .
figure [ figcap4 ] shows four images corresponding to the _ hst_/acs field of view : h i column density ( see paper i ) , _ hst _ v - band , _ galex _ near - uv @xcite , and continuum - subtracted h@xmath17 ( see paper i and kennicutt etal .
2008 ) . overlaid on each panel is the h i column density contour at the 10@xmath12 @xmath13 level .
the largest h i column density ( in the sw ) is coincident with a high surface brightness stellar cluster in the uv image , as well as with the brightest h@xmath17-emitting region .
the h i peak in the se region also has coincident uv and diffuse h@xmath17 emission .
the northern h i feature is coincident with a compact h ii region and with a stellar association visible in both the _ hst _ and the _ galex _ images .
paper i identified eight holes in the h i distribution of ddo165 using radius - velocity and/or position - velocity analyses .
figure [ figcap4 ] shows six of those regions in blue , using the same numbering scheme as paper i ( holes 3 and 8 from paper i fall completely or partially outside the _ hst_/acs field of view ) ; table [ t1 ] summarizes the properties of these structures from paper i. we seek to probe the sfhs of the stellar populations within these regions and to compare them with the sfhs extracted in other areas of ddo165 . for comparison , we extracted the sfhs of 23 other regions within ddo165 .
as shown in figure [ figcap4 ] , these regions sample a wide range of stellar and h i surface densities , h@xmath17 and uv surface brightnesses , and localized h i dynamics ( compare with paper i ) .
we explicitly study the stellar populations associated with the northern h i component ( shown in green in figure [ figcap4 ] as `` n1 '' and `` n2 '' , and referred to as `` northern 1 '' and `` northern 2 '' below ) and with two regions containing slightly resolved stellar clusters ( shown in cyan in figure [ figcap4 ] as `` cl 1 '' and `` cl 2 '' , and referred to as `` cluster 1 '' and `` cluster 2 '' below ) .
exterior to these regions and the h i holes , we then selected 20 control fields ( shown in red in figure [ figcap4 ] as `` c1''``c20 '' , and referred to as `` control 1 '' through `` control 20 '' below ) .
all regions except for northern 1 and northern 2 have a diameter of 20 ( the size of the h i beam used in this analysis ; 20 @xmath16 430 pc at our adopted distance ) ; this choice is necessarily a compromise between region size and number of stars in the resulting cmds .
the 20 size provides of order 1,000 stars in each cmd ( sometimes many more ) while allowing us to study a sizeable number of regions within ddo165 .
the sizes of regions northern 1 ( 12 diameter ) and northern 2 ( 24 diameter ) are guided by eye to encompass the stellar association in the _ hst _ image .
table [ t2 ] summarizes the names , positions and sizes of each of these regions .
we extracted the photometry for those stars that lie within each region ; the number of stars in each region , shown in table [ t2 ] , ranges from 261 ( northern 1 ) to 28,439 ( hole 6 ) .
these stars are used to construct the cmds of each individual region shown in figures [ figcap5 ] ( h i holes , cluster regions , northern region ) , [ figcap6 ] ( control 110 ) , and [ figcap7 ] ( control 1120 ) .
the rgb is easily identified in all panels ; the blue plume region , however , shows pronounced variations from one region to the next .
these variations imply significant variations as a function of time of the locations of active sf throughout ddo165 .
the global and local cmds described in [ s3.1 ] and [ s3.2 ]
are modeled to obtain a best - fit sfh using a sophisticated maximum likelihood approach .
we refer the reader to @xcite for details , and provide a brief synopsis of the technique here .
the stellar evolution models of @xcite are used to construct synthetic cmds , which are compared with the observed cmds using a maximum likelihood approach . given the large number of factors that contribute to the cmd of a composite stellar population ( e.g. , initial mass function , chemical evolution and sfr as functions of time , binary fraction , distance , crowding , foreground and differential extinction , and photometric completeness limits ) , certain properties are held fixed in the fitting routines . as in @xcite , for this analysis of ddo165 , the binary fraction ( 0.35 ) , the imf properties ( standard power - law , x = @xmath102.30 , from 0.1 to 100 m@xmath18 ) , the distance ( 4.47 mpc ) , the foreground extinction ( a@xmath19 = 0.080 ; schlegel etal .
1998 ) , and the photometric completeness limits ( m@xmath19 = 28.0 , m@xmath20 = 27.2 ) are held constant . the program is then allowed to search other parameters to obtain the best fit , including internal extinction , metallicity per time bin , and the sf rate ( sfr ) per time bin .
the metallicity is constrained to increase monotonically with time .
the results of this fitting routine provide the rate of sf and the chemical evolution as functions of time . as mentioned above ,
here we concentrate on the most recent 500 myr , during which we have maximum temporal resolution via the well - populated bheb sequence , and during which the chemical evolution of ddo165 is minimal ( e.g. , weisz etal .
2008 finds a change of 0.1 dex in metallicity between 0 and 1 gyr ) .
errors on the sfr as a function of time are quantified by adding the systematic uncertainties from the stellar evolution isochrones and the statistical uncertainties from monte carlo tests in quadrature .
extensive discussion of the error budgets can be found in @xcite and @xcite . applying the cmd fitting routines described above to the global cmd shown in figure [ figcap2 ] yields the sfh shown in figure [ figcap8 ] .
note that the time binning is nonlinear , with finer temporal resolution available for young stars in the upper bheb and ms .
the errors plotted include both systematic and statistical uncertainties .
as noted above and discussed in detail in @xcite , ddo165 hosts one of the longest - duration starbursts known to date ( @xmath141 gyr ) .
figure [ figcap8 ] verifies this temporally extended burst by comparing the recent sfr with sfr@xmath21 ( dashed line ) , the average sfr of the entire galaxy averaged from 0 to 14 gyr .
note that the sfr is higher throughout the most recent 500 myr compared to the lifetime average sfr , indicating an ongoing starburst event .
certain interesting features of the global sfh are apparent .
first , the sfr varies by a factor of @xmath02 during the last 500 myr : during the intervals from @xmath0100 myr to 25 myr , and again between @xmath0400 and 300 myr , the sfr is significantly higher than the average over the last 500 myr .
second , the sfr drops significantly between 25 myr and the present time ; this is in agreement with the lack of bright h@xmath17 emission in ddo165 ( see discussion in paper i and figure [ figcap4 ] ) .
the stars that populate these recent time bins are the most luminous bheb and ms stars ; if they were present in large numbers in ddo165 , they would be prominent in the global cmd .
the lower sfr over the last 25 myr is thus interpreted as a significant decrease in the total sfr in ddo165 .
whether this current drop in sfr corresponds to the ultimate truncation of the gyr - length burst is not discernible from these data .
figures [ figcap9 ] , [ figcap10 ] , and [ figcap11 ] present the sfhs of the selected regions in ddo165 shown in figure [ figcap4 ] and discussed in [ s3.2 ] ; each region is normalized to show sfr per unit time per unit area .
the normalized recent sfhs span an intensity range of @xmath03 over the last 500 myr in these regions .
there are clear variations in normalized sfrs from one field to the next : some regions have constant sfrs over the last 500 myr ( e.g. , hole 1 , control 5 ) , while others have dramatic variations in their sfrs during certain intervals ( e.g. , northern 1 , hole 2 , cluster 2 , control 12 , control 13 ) . comparing with the multiwavelength properties shown in figure [ figcap4 ] , the localized sfhs within ddo165 arise from a wide variety of local conditions .
two important time intervals are apparent from the localized sfhs : 400 - 300 myr and 100 - 25 myr .
these epochs are identified as periods of enhanced local normalized sfr compared to the 500 myr average . in the more ancient event , the global sfr increases by @xmath050% compared to the 500 myr average .
a corresponding increase in normalized sfr is seen in h i holes 1 and 6 , cluster 1 , and control 1 , 2 , 5 , 8 , 9 , 11 , 12 , 13 , and 17 .
these regions are distributed over a large fraction of the stellar body of ddo165 , indicating that a global sf event took place during this period .
the second period of enhanced sf is 100 - 25 myr ; the global sfr again increases by @xmath050% .
a corresponding elevated sfr is seen in regions located throughout ddo165 : holes 2 , 5 , and 6 , clusters 1 and 2 , the northern h i component ( northern 1 and northern 2 ) , and control 1 , 3 , 5 , 7 , 8 , 10 , 12 , 13 , 17 , 18 and 19 .
the fact that this more recent sf event occurred in widespread regions of the galaxy ( compare , for example , the locations and sfhs of northern 1 , control 1 and 18 ) suggests a second global sf event .
the rapid truncation of the global sfr , as well as the decrease in most normalized sfrs , suggests that this event ended as quickly as it began .
we note with interest that the 400 - 300 myr interval identified in this sfh analysis agrees with two other relevant timescales for ddo165 .
the first is the well - known three - body tidal interaction between m81 , m82 , and ngc3077 that occurred @xmath0300 myr ago @xcite .
however , since no direct evidence is seen in our h i data for a recent tidal interaction ( see paper i ) , nor does any kinematic model of the m81 group suggest that ddo165 was directly involved in the m81 - m82 - ngc3077 interaction ( e.g. , * ? ? ?
* ) , this timescale may be coincidence .
the second and perhaps more important timescale is t@xmath22 , the time required to remove evidence of correlated stellar structure .
@xcite has analyzed the evolution of stellar structures in various m81 group dwarf galaxies , and finds that stellar structures survive for at least 350 myr in ddo165 .
the long t@xmath22 for ddo165 ( compare to @xmath1480 myr for the smc and @xmath14100 myr for ngc2366 ; see gieles etal .
2008 and bastian etal .
2011 ) guarantees that the recent sfhs within the various regions of the galaxy can be reliably compared with the current h i properties of the same regions ; we discuss this point further in [ s4.3 ] below .
we have identified 20 individual stellar clusters to examine any temporal consistencies with the sfhs of the fields presented so far .
the clusters were identified by inspection of a f555w / f814w color _ hst _ image .
the masses and ages of the clusters tabulated in table [ t3 ] were determined from fitting broadband bvri ground - based photometry to single stellar population models @xcite .
the bvri data are part of the spitzer infrared nearby galaxies survey ( sings ; kennicutt etal .
2003 ) ; images were acquired with the _ kpno 2.1-m _ telescope .
figure [ figcap12 ] shows the spatial relationship of these individual clusters to the h i holes and cluster regions identified above .
note that we use lower case `` c '' labels for individual clusters in table [ t3 ] .
holes 5 and 6 both contain stellar clusters , as do the regions cluster 1 and cluster 2 .
conversely , there are 4 holes that do not contain readily identifiable clusters . also of interest
is that the strongest emission in h@xmath17 and uv comes from the region inhabited by the individual clusters c2 , c5 , c6 , and c7 .
next we examine the cluster age properties in relation to the global sfh of ddo165 over the last 500 myr .
figure [ figcap13 ] shows the global sfh ( identical to the sfh shown in figure [ figcap8 ] ) , overplotted with the masses and ages of individual stellar clusters .
the most notable feature is the temporal concentration of clusters ( c7 , c9 , c10 , c13 , c16 , c17 , and c19 ) near 100 myr .
this age corresponds to the increase in global sfr at the onset of the 100 - 25 myr sf event discussed above .
note also that the 7 clusters with ages of @xmath0100 myr are scattered throughout the entire disk of ddo165 .
taken together , these properties suggest that the formation of stellar clusters is directly linked to the increase in global sfr in ddo165 . based on the results for ddo165 presented here , the connection between h i holes and stellar clusters appears to be somewhat ambiguous : h i structures are identified both with and without interior stellar clusters .
however , it is important to note that all of the observed clusters have ages @xmath1190 myr ( see table [ t3 ] ) .
this suggests that we are observing only those associations that have survived the well - documented `` infant mortality '' phase ( e.g. , lada & lada 2003 ; fall etal .
2005 ; chandar etal . 2006 ; and references therein ) .
if the clustered sf in ddo165 has proceeded in a manner similar to that observed in other systems , it seems probable that many young clusters have been destroyed within the past @xmath0100 myr .
it is interesting to note that the relation between total sfr ( clusters plus field stars ) and cluster formation rate during periods of heightened sf activity appears to be linear in log - log space @xcite . to date , this relation has been described in the literature for 10 galaxies , none of which are dwarfs .
the results above for ddo165 appear to agree with this log - log linear relationship between total sfr and cluster formation rate .
further work to solidify this realtion in low - mass galaxies would be very instructive .
the preceding discussion highlights that sf has pervaded the disk of ddo165 over the past 500 myr .
figure [ figcap3 ] shows that over this time interval , the bheb stars are distributed more or less uniformly throughout the inner h i disk .
however , note that the truncated stellar surface brightness of the southern regions ( compare with the figures presented in paper i ) remains coherent in the bheb populations at all age intervals .
this result , together with the characteristic t@xmath22 @xmath23 350 myr derived in @xcite , motivates our comparison of the patterns of sf over the last 500 myr with the current h i morphology and dynamics .
although the h i data lack temporal information , the arguments for the dynamical simplicity of dwarfs ( e.g. , lack of internal shear , long dynamical timescales ) presented in paper i , and the predicted survival timescales of h i holes from simulations ( @xmath0100 - 600 myr ; recchi & hensler 2006 ) , suggest that the comparison of recent sf and present - day h i properties are physically motivated in low - mass systems .
we take advantage of the fine temporal and physical resolution in our _ hst _ data by presenting movies of the normalized sfr .
these movies were created using a procedure similar to that described in @xcite and applied by @xcite ; we refer the reader to the latter work for detailed discussion .
the first step is to create a series of stellar density maps on a specified grid size .
knowing the physical coordinates and ages of the bheb stars , they are binned into 5 myr intervals and gridded in right ascension and declination .
these images are then smoothed spatially by a 17 gaussian kernel ( most directly comparable to the 20 h i beam size from paper i ) and then normalized by dividing by the total number of counts .
the second step is to create a sfh on the same temporal grid .
the global sfh is binned with different time resolutions to account for the degraded resolution with sfh look back time : 10 myr for the 0 - 100 myr interval , 20 myr for the 0 - 200 myr interval , and 50 myr for the 0 - 500 myr interval .
these are linearly interpolated onto the same temporal grid as the stellar density maps ( 5 myr ) . the multiplicative product of the stellar density maps and the sfh is divided by the physical area of the field of view , thus resulting in a series of images representing sfr per unit area , each separated by 5 myr .
the resulting movies ( available in the electronic version of the journal ) show the intensity of sf as a function of time and of position , to lookback times of 100 myr , 200 myr , and 500 myr , respectively .
note that while each movie has the same time step ( 5 myr ) , the movies extending to larger lookback times were created by interpolation over larger time intervals and thus have a higher relative uncertainty from one time step to the next . for clarity ,
each movie contains the same h i 10@xmath12 @xmath13 column density contour as shown in figure [ figcap4 ] .
figures [ figcap14 ] , [ figcap15 ] and [ figcap16 ] show selected still frames from the movies .
it is important to emphasize that these movies and images show a sfr per unit area ; high sfrs can arise from low - level sf distributed over large areas , or from high - level sf in a small region . using the heightened period of sf around 100 myr
( see figure [ figcap8 ] ) as a guide , the sfrs per unit area in the movies near this lookback time are at modest levels but occur over a substantial portion of the system .
the movies highlight various important features of the recent evolution of this intriguing system .
first , the starburst in ddo165 is a global event ; throughout the last 500 myr the sf has been ongoing throughout most of the stellar component .
second , the period of heightened global sfr rate from 400 - 300 myr is apparent as distributed sf throughout the system and centered on the current stellar distribution ; the elevated global sfr from 100 - 25 myr is manifested as somewhat more localized regions of sf ( preferentially located in the southern region ) superposed on distributed lower - level sf .
third , there is little sf exterior to the 10@xmath12 @xmath13 column density contour .
this is especially evident in the southern region as the aforementioned truncation of the stellar population . finally , comparing with figure [ figcap3 ]
shows that sf inside the giant h i hole has spanned the entire 500 myr interval ( although only a small number of stars with ages @xmath2425 myr are present ) .
in addition to the h i holes and shells discussed above , paper i identified a wealth of high - velocity gas within ddo165 .
most of this gas is located inside of the southern 10@xmath12 @xmath13 contours shown in the movies and in figures [ figcap14 ] , [ figcap15 ] , and [ figcap16 ] . a careful examination of the spatial distribution of bheb stars in these regions ( figure [ figcap3 ] )
, the normalized sfhs of these regions ( e.g. , clusters 1 and 2 , hole 2 , control 7 , 9 , 12 , 13 , 18 ) , and the sfr per unit area movies shows that there has been significant sf in the southern h i component at locations coincident with the present - day high velocity neutral gas . in agreement with paper
i , extended sf events appear capable of creating incoherent but high velocity motion of the surrounding interstellar material .
we note that coherent h i structures smaller than our physical resolution limit ( @xmath0160 pc using our highest resolution h i cubes ) may be present in ddo165 .
paper i identified 8 holes and shells in the neutral ism of ddo165 .
of those 8 structures , 5 were found to have kinematic expansion signatures .
table [ t1 ] summarizes the properties of the h i holes that lie within the _ hst _ field of view .
for the three kinematically expanding structures in table [ t1 ] , paper i made the the simplistic assumption that each structure is created by a single blast . a coarse estimate of the energy required to create the expanding h i structure
was then calculated via @xmath25 where e@xmath26 is the energy needed to create the expanding structure in erg , n@xmath27 is the h i volume density before the creation of the hole ( paper i assumes n@xmath27=0.1 @xmath28 ) , d is the diameter of the shell in parsecs , and v@xmath29 is the observed expansion velocity in kms@xmath5 .
these e@xmath30 values are tabulated in table [ t4 ] ( columns 2 , 3 , and 4 ) . as discussed in paper
i , this calculation is overly simplistic for a number of reasons ; we now attempt a more realistic and physically motivated derivation of the energy requirements for the observed h i structures by modeling the mass surface density and the h i volume density . for kinematically expanding structures ,
the velocity and physical size are easily measured ; the volume density , however , is difficult to measure directly and we thus rely on a mass model approach to obtain the behavior of n@xmath27 as a function of position . for a disk of h i gas , the column density ( n@xmath31 ) , volume density ( n@xmath27 ) and disk scale height ( _ h _ )
are related via @xmath32 since the velocity field of ddo165 is so peculiar , paper i was not able to achieve an unambiguous rotation curve fit , which could have been used to determine the scale height of the galaxy .
we thus rely on infrared photometry to produce a stellar mass model of ddo 165 .
note that such a model implicitly assumes that the bulk of the infrared luminosity arises from the stellar populations older than the ms or bheb stars ; this population is assumed to retain the same mass distribution both before and after a sf event .
we used the irac 3.6 @xmath33 m images acquired with the _
spitzer space telescope _ as part of sings ( kennicutt etal .
2003 ; see also walter etal .
2007 ) to derive our stellar mass model ( see discussion in oh etal .
2008 and application in weisz etal .
after removing obvious foreground stars , the 3.6 @xmath33 m image ( which also shows the truncation of the stellar population in the southern region of the system ; see figure 4 of walter etal .
2007 ) was spatially smoothed to 7.5 resolution ( our highest h i angular resolution element ) and then fitted with ellipses of constant surface brightness .
the best - fit solution yielded a central position of ( @xmath17,@xmath34 ) = ( 13:06:24.7 , 67:42:33.2 ) , an ellipticity of 0.5 , and a position angle of 90@xmath35 .
the radially averaged 3.6 @xmath33 m surface brightness profile is shown in figure [ figcap17 ] ; the surface brightness declines with radius , falling from @xmath021 magarcsec@xmath36 to @xmath023 magarcsec@xmath36 within @xmath03 kpc .
the h i scale height of an isothermal disk is related to the velocity dispersion of the gas and the mass surface density . as derived in @xcite and applied in @xcite and @xcite , @xmath37 where @xmath38 is the h i velocity dispersion in kms@xmath5 ,
@xmath39 is the mass density profile in m@xmath18pc@xmath36 , and _ h _ is the h i disk scale height in pc . using the same central position ,
ellipticity and position angle used to derive the stellar surface brightness profile , we extracted radially averaged values of h i column density and velocity dispersion .
these plots are shown in figure [ figcap17 ] .
the rise in h i velocity dispersion in the central @xmath00.5 kpc , and the prominent low column densities in the same region , are the results of the giant h i hole .
we note that the observed velocity dispersions may not be appropriate for calculating the scale height of the gas prior to the onset of sf ; we thus also perform energy calculations with assumed velocity dispersions similar to those seen in the quiescent regions of nearby dwarf galaxies ( see further detailed discussion below ) .
figure [ figcap18 ] shows radial profiles of derived properties derived from observed properties shown in figure [ figcap17 ] and using the equations above .
the top panel of figure [ figcap18 ] shows the mass surface density in units of m@xmath18pc@xmath36 , derived from the 3.6 @xmath33 m surface brightness profile by assuming a 3.6 @xmath33 m mass to light ratio of 0.5 ( similar to other dirr systems ; see de blok etal .
2008 ) and applying the techniques used in @xcite and @xcite .
the middle panel of figure [ figcap18 ] shows the h i scale height , _ h _ , as a function of radius , derived using the radial values of @xmath40 and @xmath39 .
the shape of the h i scale height as a function of radius is qualitatively similar to the profile derived for the m81 group dwarf holmbergii in @xcite using a similar approach ; however , the scale height in the outer disk of ddo165 is a factor of @xmath03 larger than for holmbergii .
the red line shows the polynomial fit to this curve , which we use to obtain the value of _ h _ at any given radius .
finally , the bottom panel of figure [ figcap18 ] shows the midplane h i volume density as a function of radius , derived using the radial values of n@xmath31 and _ h_. the giant h i hole is prominent at radii @xmath241 kpc . moving outward toward r @xmath16 1 kpc ,
the scale height remains essentially constant , @xmath39 is decreasing , @xmath38 falls to a more or less constant value , and the h i column density rises .
these properties conspire to create a maximum in the h i volume density at r @xmath0 1 kpc .
the volume density then falls smoothly moving further outward through the disk ; the increasing scale height implies lower h i volume densities at these large radii .
three fits to the radial plot of h i volume density are shown in figure [ figcap18 ] .
the red line shows the functional fit including all data points . since there was h i gas inside this structure when the bheb stars were formed , our derivation of the energy requirements for the formation of this hole would be an underestimate if we used the present - day volume density .
thus , we include two other fits to the n@xmath27 curve : the function shown in blue is fitted to the data beyond 1 kpc and is forced to plateau to a constant volume density in the inner disk ; the function in green is also fitted to the data beyond 1 kpc using a gaussian function that rises to slightly larger n@xmath27 values in the inner disk . while these fits do not explicitly conserve h i mass , we stress that the volume density enters the energy calculation as n@xmath41 ; the choice between these three curves produces significant differences in energy only in the inner regions of the galaxy ( i.e. , where the red and blue / green curves differ substantially ) . the radial behavior of h i volume density now allows us to estimate the creation energies for the h i holes .
we begin by considering those 3 holes with measured expansion velocities ( 2 , 5 , 7 ) . knowing these velocities and the sizes of the holes ( radius for circular structures , geometric radius for elliptical structures ) , we simply insert the volume density values at the appropriate galactocentric radius ( figure [ figcap18 ] ) into equation [ eq1 ] .
the results are shown in table [ t4 ] ; for each hole , we estimate the requisite creation energy assuming four values of the h i volume density . the simplistic constant n@xmath27 approach ( as in paper i )
is shown in columns 3 and 4 ; the polynomial fit approach is shown in columns 5 and 6 ; the best fit line at r @xmath14 1 kpc and plateau to constant n@xmath42 at r = 0 approach is shown in columns 7 and 8 ; the best fit gaussian profile at r @xmath14 1 kpc approach is shown in columns 9 and 10 .
three important insights arise from an examination of table [ t4 ] and figure [ figcap18 ] .
the first is that the most significant differences in volume density occur between the simplistic n@xmath27 = 0.1 @xmath13 assumption ( close to the value at r @xmath23 1 kpc ) and the outer disk regions .
this implies that estimates of the energy requirements for a hole in the outer disk will be significantly lower using the mass model approach discussed above than using the assumed value from paper i. hole 7 shows this effect quite clearly , as the requisite energies differ by a factor of two .
the second insight is that for all holes at radii beyond @xmath01 kpc , the functional fits in figure [ figcap18 ] use the same data and thus produce essentially identical n@xmath27 values and energy requirements .
the final important feature of table [ t4 ] is that the creation of each of these kinematically expanding structures requires of order 10@xmath2 erg .
this estimate is similar to the energy requirements for the smaller holes in holmbergii found by @xcite .
we next consider those holes that show no signs of kinematic expansion , rendering a direct application of equation [ eq1 ] difficult .
these kinematically static structures are usually assumed to have either blown out of the disk or to have re - established pressure equilibrium with the surrounding ism .
the latter effect will occur when the average bulk velocity of a shell equals the average velocity dispersion over a given region .
thus , if we assume coherent motion at an average velocity equal to that velocity dispersion , we can estimate the energy requirements for the structure .
we stress that such estimates are uncertain at the 50% level or more ; nonetheless these numbers are useful when compared to the mechanical energy injected by the evolving massive star populations ( see below ) . in an attempt to bracket the potential extrema in the assumed velocity ranges
, we consider three different values for the putative expansion velocities of the kinematically stalled structures .
first , we assume a value of 7 kms@xmath5 for any stalled hole , as adopted by the analyses in the hi nearby galaxies survey ( things ; walter etal .
next , we can adopt one of two different values for the velocity dispersion : @xmath38 averaged over the entire hole structure , or @xmath38 averaged over the entire galaxy . as expected , the global @xmath38 value ( [email protected] kms@xmath5 , regardless of angular resolution ) is lower than in localized regions ( @xmath38 @xmath23 11 - 12 kms@xmath5 ) .
the energy estimates for the stalled structures are summarized in table [ t5 ] .
column 2 gives the assumed expansion velocity for each hole ; the remaining columns contain the same information as those in table [ t4 ] : constant n@xmath27 in columns 3 and 4 ; the best - fit polynomial function in columns 5 and 6 ; the best fit function plateaus to constant n@xmath27 in columns 7 and 8 ; the best fit gaussian in columns 9 and 10 .
the table repeats these calculations for each assumed v@xmath43 value .
similar trends are seen for these stalled holes as described above : the largest differences at a chosen velocity occur between holes in the outer disk and holes at the maximum n@xmath27 value ; energy estimates for holes in the outer disk are independent of the adopted n@xmath27 value .
further , for the smaller h i holes in table [ t5 ] ( 1 , 4 ) the estimated creation energies are similar to those found for the expanding features in table [ t4 ] . note by comparing the energy estimates for a given hole that the v@xmath44 dependence results in a factor of @xmath02 variance in the derived energy budget .
the giant h i hole ( hole 6 ) stands out prominently in table [ t5 ] .
as expected , the characteristic creation energy of this structure is roughly two orders of magnitude larger than the other holes .
further , this characteristic energy of @xmath010@xmath45 erg is similar to those derived for other giant h i structures in dwarf galaxies ( e.g. , walter & brinks 1999 ; weisz etal . 2009b ; warren etal .
2011 ) . the information in tables [ t4 ] and [ t5 ] allows us to estimate kinematic ages for each of these structures ;
these ages are given in table [ t6 ] .
for the expanding holes ( table [ t4 ] ) this number is simply the geometrical mean radius divided by the measured expansion velocity ( assumed to be constant ) .
note that these ages are explicitly upper limits since the expansion velocity was higher immediately after the hole was created .
for the stalled structures , we adopt the global velocity dispersion as being representative of the current expansion velocity and use it in subsequent discussion .
this is motivated by two considerations .
first , the global @xmath40 value provides energy estimates near the middle of the ranges shown in table [ t5 ] .
second , the galaxy - averaged velocity dispersion sets the characteristic ism pressure and thus determines in large part when an expanding h i structure will stall .
the kinematic ages tabulated in table [ t6 ] range from 22 to 89 myr .
this time interval corresponds to the period of heightened sfr between 25 - 100 myr discussed above ; it also sets the characteristic timescale of 100 myr over which we will compare the energy outputs from stellar evolution with the energy requirements from the h i data . in the discussion that follows
we adopt the energy estimates derived using the gaussian profile for h i volume density of all h i holes . for expanding holes
, v@xmath43 is an observed quantity . as justified above for the stalled holes , we hereafter set v@xmath43 @xmath16 8.5 kms@xmath5 , i.e. , equivalent to @xmath38 for the entire galaxy .
these assumptions lead to the e@xmath46 values in column 3 of table [ t6 ] .
while the relative errors on these values are difficult to estimate , we stress that our calculations should be considered uncertain at the 50% level ( at minimum ) ; the scatter in e@xmath46 estimates from tables [ t4 ] and [ t5 ] can also be used to guide interpretation .
we emphasize that the specific e@xmath46 values are less important than the order of magnitude estimates ; [ s4.3 ]
shows that the feedback energies from the stellar populations inside each hole exceed these e@xmath46 estimates , usually by more than an order of magnitude .
the global and localized sfhs presented above allow us to calculate the integrated mass of the stars formed as a function of time ; this is simply the product of average sfr and time bin width . with the assumption of a fully populated universal salpeter imf ,
this total stellar mass implies a certain number of massive stars that will result in sne and therefore deposit mechanical energy into the surrounding ism .
the amount of feedback energy resulting from the evolution of these massive stars can thus be quantified using our sfhs and models of the energy outputs of sf episodes . to this end
, we apply the starburst99 models @xcite to each of the sfhs described in [ s3 ] above .
specifically , we perform a simulation of a 10@xmath47 m@xmath18 instantaneous burst of sf , scaled to the integrated mass in each sfh , using the same relevant parameters as discussed in [ s3.3 ] .
it is assumed that each star more massive than 8 m@xmath18 explodes as a sn @xcite .
we make no correction for the contributions of type ia sne to the feedback energy budget .
these events occur over a range of timescales @xcite and may occur in regions that are already partially evacuated ; for both reasons , the actual feedback energy from stellar evolution in a given region can be considered to be a lower limit . the starburst99 models are sampled every 5 myr in order to create profiles of the feedback energy resulting from sf as a function of time .
our _ hst _ imaging provides detailed information on the energy budget as a function of time ; however , we are cognizant of the uncertainties introduced as a result of random stellar diffusion . as stars move within a galaxy , they will disperse away from their formation sites ; this effect becomes important as stars venture into and out of given areas of consideration within a galaxy .
while the results presented by @xcite suggest a long period of coherence of stellar structure within ddo165 ( see discussion above ) , we conservatively limit our consideration of the input energies within a given region to 100 myr . by considering only this recent interval
, we can minimize the uncertainties resulting from stars diffusing into or out of a given region . by way of example , the control fields have physical radii of @xmath0220 pc at the adopted distance .
a 2 kms@xmath5 net velocity will move this distance in @xmath0100 myr ; however , the relaxation dynamics are likely complex and the actual crossing time will depend on many paramaters ( local stellar and gas density , burst mass and age , etc . ) . for the larger h i holes we observe in ddo165 ( especially # 6 , with a minor axis of @xmath01.1 kpc )
, the effects of diffusion out of the region will be smaller yet .
the results of our energy calculations over the most recent 100 myr are shown in figures figures [ figcap19 ] , [ figcap20 ] , and [ figcap21 ] .
a comparison of these profiles with the normalized sfh plots ( figures [ figcap9 ] through [ figcap11 ] ) demonstrates several interesting points .
first , total feedback energies over the last 100 myr fall in the @xmath010@xmath210@xmath48 erg range .
further , the energy input into each of these regions is temporally extended ; the canonical assumption of an instantaneous injection of feedback energy during the formation of a stellar cluster is a simplification compared to the extended energy profiles .
note that these profiles could be extended to larger lookback times ( with appropriate caveats ) , increasing the total energy feedback budgets .
the cumulative energy profiles shown in figures [ figcap19 ] through [ figcap21 ] allow us to quantify the amount of energy resulting from recent sf on any timescale of interest between 0 and 100 myr .
these energies can then be directly compared with the energy requirements for creating the h i structures ( see discussion in [ s4.2 ] above ) .
recall that table [ t6 ] shows the adopted kinematic ages of the h i structures ( column 2 ) , as well as the requisite energies to create them ( column 3 ) .
it is now straightforward to find the amount of energy produced by sf and stellar evolution over the adopted kinematic ages of the structures : one simply reads the value of energy at the appropriate age in the panels of figure [ figcap19 ] . to guide interpretation
, that figure shows the kinematic age of the structure by a vertical line .
column 4 of table [ t6 ] tabulates e@xmath49 , the total energy input by stellar evolution over the adopted kinematic age of the h i structure .
note by examining the sfhs and energy profiles of h i holes 4 ( stalled structure ) and 7 ( expanding structure ) that an estimate of e@xmath49 is not possible for these features : no recent sf has occurred within the kinematic ages of these holes .
this might not be surprising for hole 4 : it is located in a region of low h i column density and low stellar density .
hole 7 is much more difficult to interpret ; it contains dense neutral gas ( overlapping the 10@xmath12 @xmath13 contour ) and is expanding at 9 kms@xmath5 .
however , it hosts only low - level sf over the past 500 myr . for the other four h i structures ,
the energy derived from sf over the kinematic age is larger than the energy required to create the structure .
we term the ratio e@xmath46/e@xmath49 the `` feedback efficiency '' .
the efficiency is relatively high for hole 1 ( @xmath040% ) but low for the giant h i hole ( @xmath01% ) ; these efficiencies are similar to those found for holes in holmbergii by @xcite . for these h i structures
, recent stellar evolution provides sufficient energies within the kinematic ages of the hole in order to create them .
we note that e@xmath49 @xmath14 e@xmath46 for any of the energy estimates in tables [ t4 ] or [ t5 ] , even when selecting the highest e@xmath46 values .
the assumption of 100% efficiency is implicit ; smaller values of efficiency produce smaller energy excesses , but retain sufficient cumulative energy inputs to create the observed structures .
as noted above , we can only estimate the kinematic ages of structures that are expanding at the present time .
as such , a second timescale of interest is the amount of time required for the energy from sf to equal the energy required to create the structure .
we hereafter refer to this characteristic timescale as t@xmath50 , the time when e@xmath49 @xmath16 e@xmath46 ( assuming 100% efficiency ) ; we tabulate the values of t@xmath50 in column 6 of table [ t6 ] .
this time interval is very brief for those structures with sf in the last @xmath050 myr : the time required is less than 30 myr in all cases .
for those structures without recent sf ( holes 4 and 7 ) , this interval is much longer : @xmath0130 myr for hole 7 and @xmath0330 myr for hole 4 .
these extended intervals should be treated with caution ; as noted above , on timescales longer than 100 myr , random stellar motions of a few kms@xmath5 can move stars both into and out of areas of the size we consider here .
we can envision a second and related source of uncertainty in the interpretation of the energy timescales .
consider a concentrated region of sf that imparts energy into the surrounding ism and creates an expanding structure ; as the hole diameter increases , it can cover a much larger area than the sf that created it . if there were any recent sf in this surrounding area , then the energy budget increases , although that energy did not specifically contribute to the creation of the h i hole . in this sense
the energy estimates of our structures may be overestimates .
however , it seems likely that large contributions from concentrated sf would create h i structures of their own ; as the two structures merge they create a single structure .
further , we make no corrections for the energy contributions from type ia sne ( see discussion above ) , arguing for the derived energies being underestimated . with these uncertainties in mind for holes 4 and 7
, we note that t@xmath50 is insensitive to the adopted e@xmath46 value : again selecting the largest e@xmath51 values from tables [ t4 ] and [ t5 ] only changes t@xmath52 for holes 1 ( t@xmath50 @xmath24 35 myr ) , 4 ( t@xmath50 @xmath24 400 myr ) , and 6 ( t@xmath50 @xmath24 25 myr ) .
a direct comparison of the energies from sf with the energies of the surrounding gas is by definition limited to the h i holes only .
however , following the discussion above , we can consider the feedback energies injected over only the most recent @xmath0100 myr in any region within ddo165 .
we thus tabulate in table [ t7 ] the feedback energies from sf after 25 , 50 and 100 myr in each of the regions shown in figure [ figcap4 ] .
this allows a comparison of the energies from recent sf in different regions throughout the galaxy in a straightforward manner .
multiple interesting trends are seen in the data presented in table [ t7 ] .
first , most regions have accumulated of order 10@xmath53 erg of energy by 25 myr and of order 10@xmath45 erg of energy by 100 myr .
this enforces the importance of temporally extended sf on the surrounding ism ( e.g. , * ? ? ?
second , the regions with ongoing sf within the most recent intervals stand out prominently : for example , cluster2 has the largest e@xmath49 value at 25 myr ( 2.8@xmath110@xmath45 erg ) , exceeding even the feedback energy within the much larger giant h i hole .
remarkably , this substantial energy budget has failed to produce a coherent h i structure at our spatial and velocity resolution ( see further discussion below ) .
third , the increase in the cumulative e@xmath49 values between 25 and 100 myr ( when e@xmath49 at 25 myr is not zero ) is at least an order of magnitude for many ( but not all ) regions .
this again suggests that the stellar - ism interaction operates over timescales much longer than the lifetimes of massive stars .
finally , and perhaps most importantly , the control fields have similar energy budgets on these timescales as do the h i holes .
taken at face value , the data in table [ t7 ] suggest that abundant feedback energy is injected throughout the disk of ddo165 ; coherent h i structures form in some regions but not others .
the above discussion has demonstrated that significant feedback energy is produced by stellar evolution processes in regions throughout ddo165 that span a large range of physical parameter space . in the discussion that follows , we discuss the stellar feedback hypothesis as applied to both the h i holes and the regions located throughout the system . while alternatives to the feedback hypothesis do exist ( e.g. , high velocity cloud impacts , disk instabilities , turbulence , ram pressure stripping ; snchez - salcedo 2002 and references therein ) , our data are optimal for examining the energies of stellar evolution and of h i structures .
multiple recent works have applied similar techniques in studying the feedback mechanism in nearby galaxies ( e.g. , * ? ? ?
* ; * ? ? ?
* ) ; we refer the reader to the references in those works ( as well as in paper i ) concerning potential alternative creation mechanisms .
we now address a deceivingly simple question : are the observed h i holes the results of sf that occurred in the past 100 myr ?
we consider each structure individually to address this question .
three holes are stalled ( 1 , 4 , 6 ) and three are expanding ( 2 , 5 , 7 ) .
hole 2 can be considered the clearest example of feedback energy creating an h i hole in ddo165 .
the high stellar density and dramatic increase in sfr during the 100 - 25 myr interval produce a rapid deposit of mechanical energy into the surrounding ism .
the result is the most rapidly expanding structure identified in this study
. holes 4 and 7 have no sf within the most recent @xmath0100 myr , and thus represent challenges to the hypothesis of creation via sf feedback .
sufficient stellar feedback energies are only injected over longer timescales ( t@xmath50 @xmath23 330 myr for hole 4 and @xmath23130 myr for hole 7 ; see table [ t6 ] ) .
if these structures can remain coherent over timescales of hundreds of myr ( e.g. , recchi & hensler 2006 ) , then the creation of these structures by stellar evolution processes poses no apparent difficulties .
the location of hole 4 in the outer disk and the low surrounding h i surface densities might argue in favor of such a long - lived coherent structure .
however , the location of hole 7 poses a significant issue : given the global velocity dispersion value ( @xmath38 @xmath16 8.5 kms@xmath5 ) , the crossing time of that structure is of order 25 myr .
this implies that the surrounding gas will re - establish pressure equilibrium on similar timescales .
further , the expanding nature of the structure is challenging to reconcile with the lack of recent sf over the most recent @xmath0100 myr .
we interpret these structures as evidence against a recent ( @xmath4100 myr ) feedback origin ; however , feedback on longer timescales remains a viable mechanism .
holes 1 and 5 present a similar juxtaposition ; 5 is expanding while 1 is stalled . both structures
have e@xmath49 @xmath14 e@xmath46 , implying that sf is energetically capable of producing the holes .
hole 1 has a higher normalized sfr level and a higher stellar density than hole 5 .
hole 5 , on the other hand , has experienced a recent period of slightly elevated sfr ( 100 - 25 myr ) ; this can be interpreted as a potential link between recent sf and hole creation .
the giant h i hole is perhaps the most challenging structure to interpret in ddo165 .
the energies derived from sf are clearly energetically capable of removing the h i gas from within . however , the size of the structure , the low column densities within , and the presence of recent sf cloud this simplistic interpretation . in order for stars to have formed recently
, the structure must either have retained interior local h i overdensities beyond our resolution limit or have expanded to its present size extremely rapidly ( within @xmath050 myr , implying v@xmath43 @xmath2325 kms@xmath5 , easily detectable at our velocity resolution ) .
alternatively , it is possible that the elevated sfr in the 400 - 300 myr interval caused the ism to become highly fragmented ( such as in ic2574 and hoii ; see weisz etal . 2009a , 2009b )
; the feedback energy injected by the more recent sf then fully evacuated the remaining gas . taken as a group ,
we conclude that the coherent h i holes identified in paper i can be created by stellar evolution processes , so long as the structures can remain coherent over 100 myr timescales
. stated equivalently , our data are in agreement with a model where a mixed age and spatially distributed stellar population produces sne separated by tens or hundreds of myr .
the cumulative energy outputs of these populations of stars are responsible for the creation of the h i holes .
this is in agreement with the conclusions of @xcite based on a larger sample of holes in the m81 group dwarf holmbergii , as well as with the conclusions of @xcite for a collection of dirr galaxies with giant h i holes .
we now consider the origin of the kinematic discontinuity between the northern and southern h i components .
as paper i shows , these components are static with respect to one another along the line of sight .
is this separation the result of macroscopic motion of h i gas due to sf ( i.e. , outflow ) or is it the pristine infall of an h i cloud ?
paper i uses arguments based on the h i data alone to eliminate various potential origins for the northern cloud ( including superposed or counter - rotating disks , a coherently rotating disk , and turbulence ) .
however , based on the h i data alone , we can not differentiate between infall and outflow origins for the northern component .
our energy analysis based on the recent sfh provides tentative evidence for the outflow scenario . as shown in figure
[ figcap22 ] , within the last 500 myr ( roughly half of the burst duration ; see mcquinn etal .
2010 ) , the stars within ddo165 have injected @xmath010@xmath54 erg of energy into the ism .
this cumulative energy budget is in principle capable of moving the @xmath01.65@xmath110@xmath55 m@xmath18 associated with the northern component ( see paper i ) @xmath01 kpc out of the total gravitational well of ddo165 .
we estimate the required energy via a simple gravitational potential energy calculation : @xmath56 where 8.56@xmath110@xmath57 is a constant that incorporates g and conversion factors , m@xmath58 is the mass of the northern h i component in m@xmath18 , m@xmath59 is the total mass of the ddo165 system in m@xmath18 , and r is the separation between the northern h i component and the center of the gravitational potential well . assuming m@xmath60 1.65@xmath110@xmath61 m@xmath18 , m@xmath62 10@xmath63 m@xmath18 ( i.e. , @xmath010 times larger than the total h i mass ) , and r @xmath16 1 kpc , this yields e @xmath23 10@xmath54 erg .
we stress , however , that this calculation requires the assumption of various parameters that are not observationally constrained ; some of the most important will be total depth and shape of the gravitational potential , and potential relative motion of the northern and southern components out of our line of sight ( recall that the inclination of the system remains unconstrained ) or at projected velocities lower than our spectral resolution limit .
our stellar energies thus favor the outflow hypothesis , but we treat this conclusion with caution . the outflow interpretation is strengthened considerably by the nebular spectroscopy presented in @xcite .
there , two slit positions targeted the highest surface brightness h@xmath17 regions ; these positions ( [ 13:06:16.42 , @xmath6467:42:08.4 ] and [ 13:06:24.52 , @xmath6467:43:41.2 ] ) correspond to cluster 2 in the main body and northern 1/northern 2 , respectively .
the derived oxygen abundances of these two regions are identical within the errors ( 12 @xmath64 log(o / h ) @xmath16 [email protected] and [email protected] ) .
this supports the outflow interpretation based on the well - known chemical uniformity seen in most dwarf galaxies ( see kobulnicky & skillman 1996 , 1997 , kobulnicky etal . 1997 , and references therein ) .
if the northern cloud were pristine infalling material , it would be highly unlikely to share the same gas - phase metal abundance as the southern h i component .
if the northern component became kinematically distinct as a result of sf ongoing over the past 500 myr , then the identical abundances are to be expected . while the evidence in [ s5.1.1 ] argues in favor of a stellar feedback origin for the h i holes , we are left with an equally challenging question : why are coherent h i structures created in some regions , but not in others with essentially identical physical characteristics ?
stated equivalently , why do we not observe h i holes associated with some of the more active regions studied here ?
consider , in this context , a comparison of hole 2 and control 13 .
they share similar average h i column densities , uv surface brightnesses , and stellar densities .
their cmds are similar , with strong bheb and rheb loci ; their sfhs share the same qualitative shapes and normalized sfr values .
they both have elevated sfrs in the 100 - 25 myr interval .
while the recent sf episode is stronger in hole 2 than in control 13 , hole 2 covers a slightly larger physical area .
control 13 does not create a coherent h i structure , but hole 2 does .
given the global burst of sf that ddo165 has undergone in the last 1 gyr ( see mcquinn etal .
2010 ) , and the energies available from stellar evolution processes in the regions studied here , it may in fact seem surprising that so few coherent h i structures are detected in ddo165 .
to explain this , we propose an alternative interpretation , where the ongoing sf ( which changes strength as a function of time , position and thus physical characteristics ) is capable of creating incoherent but bulk motion of the interstellar material surrounding localized sfr peaks .
this provides a simple explanation for the prevalence of the localized high velocity h i gas motions that permeate the southern h i component ( see paper i ) .
when sf is spatially concentrated and temporally extended , it is capable of creating macroscopic h i holes and shells .
when sf permeates large regions with variations in intensity ( e.g. , throughout the southern h i component , where relatively few h i holes are found compared to the number of control and cluster regions ) , the output energies serve to create incoherent but high velocity motions of the h i gas when seen at the physical resolution of these data .
this suggests that , in addition to the classical h i holes and shells , localized high velocity gas motions can be interpreted as a signature of stellar feedback .
this is in agreement with the modeling results of @xcite and @xcite , where the local velocity dispersion increases as a result of elevated sfrs .
further simulations of the interaction of spatially and extended sf on the surrounding ism , with various assumptions on the efficiency and mechanisms of coupling between the energy and the gas , would be very valuable .
we conclude that regions with significant sf activity that lack coherent h i structures are a natural product of temporally and spatially distributed sf .
this is based on multiple observed and theoretical properties of the ism .
first and most obviously , coherent h i motions smaller than our resolution limit ( @xmath0160 pc ) can not be discerned in these data .
second , spatially extended sf episodes can have both creative and destructive effects on coherent h i structures .
an examination of the movies discussed in [ s4.1 ] shows numerous examples of nearby regions of elevated normalized sf rates ; coherent motions from one may encounter opposing motions from another .
third , the porosity of the ism is challenging to measure at the present time ( e.g. , * ? ? ?
* ) and is unknown prior to the onset of sf in any given region .
feedback energy deposited into a porous region will more efficiently escape than in a uniform surrounding medium ( e.g. , * ? ? ? * ) .
finally , the position of the sf region with respect to the galactic midplane is an important parameter in the evolution of expanding structures from sf sites ( e.g. , * ? ? ?
* ) . taken as a whole
, these arguments favor a complex scenario for the deposition of feedback energy into the surrounding ism ; the efficiency of this process will depend on many factors ( e.g. , the intensity , duration and distribution of massive star formation , the metal content of the parent cloud , the local and total gravitational potential depth and shape , the local neutral and molecular gas density , the ism porosity , position with respect to galactic midplane , etc . ) , leading naturally to both coherent h i holes and to incoherent motion of the interstellar material .
we have presented a comparison of the recent sfh in ddo165 with the complex neutral gas morphology and kinematics . using _
hst _ imaging we construct cmds that allow us to isolate the age - sensitive helium burning stars .
we exploit the fine temporal resolution back to 500 myr in order to create both global and localized sfhs of selected regions within ddo165 . by comparison with synthesis models and stellar evolution isochrones ,
these sfhs can be used to derive the amount of energy unleashed by evolving stars into their immediate surroundings as a function of time .
the global sfh of ddo165 reveals interesting evolutionary characteristics .
compared to the lifetime average sfr ( 1.3@xmath110@xmath36 m@xmath18yr@xmath5 ) , the average sfr over the last 500 myr is elevated by a factor of @xmath03 .
this verifies the remarkably long burst of sf that ddo165 has undergone ( @xmath141 gyr ; see mcquinn etal .
two intervals of heightened sfr occur between look back times of 400 - 300 myr and 100 - 25 myr .
the sfr drops precipitously in the most recent 25 myr , consistent with the system s `` post - starburst '' classification @xcite .
it is important to understand how the significant feedback from this extended starburst episode has impacted the ism of ddo165 . of specific interest
are the h i holes identified in paper i ; our spatially resolved _ hst _
data allow us to examine the stellar populations within these structures and to quantify their sfhs and feedback energy budgets .
in addition to these h i holes , we study the stellar populations in 2 regions containing stellar clusters , one region associated with the kinematically distinct northern h i cloud ( with associated stellar cluster ) , and 20 control fields that sample a wide variety of optical and uv surface brightnesses , stellar densities , and h i surface densities .
many of these control fields sample regions containing high velocity gas features ( from paper i ) .
we are thus able to quantitatively examine the localized effects of massive star evolution .
the most dramatic ism feature in ddo165 is a giant h i hole that encompasses @xmath020% of all of the measured stars within the galaxy .
this structure spans @[email protected] kpc and requires @xmath010@xmath45 erg of energy in order to create it .
our _ hst _ data reveal that the stellar populations inside of this structure are energetically capable of creating this structure ; in less than 20 myr there is sufficient feedback energy to meet this requirement .
extending to a lookback time of 100 myr , there is nearly 100 times more feedback energy produced by evolving stars than are required to create this structure . applying an identical analysis to the stars within each of the other holes
leads to the same conclusion : feedback energies exceed the requisite creation energies , often by large amounts .
the other localized regions studied within ddo165 ( i.e. , the `` cluster '' , `` northern '' , and `` control '' fields ) have comparable cumulative feedback energies as the h i holes .
no coherent h i holes or shells are observed in these regions ; this complicates the stellar feedback interpretation for the creation of hi holes .
we thus propose an alternative interpretation , where ongoing sf is capable of creating incoherent but bulk motion of the interstellar material surrounding localized sfr peaks .
this provides a simple explanation for the prevalence of the localized high velocity h i gas motions that permeate the southern h i component ( see paper i ) . when sf is spatially concentrated and temporally extended , it is capable of creating macroscopic h i holes and shells . when sf permeates large regions with variations in intensity
, the output energies serve to create incoherent but high velocity motions of the h i gas .
we conclude that regions with significant sf activity that lack coherent h i structures are a natural product of temporally and spatially distributed sf . perhaps as remarkable as the giant h i hole and
the high velocity gas features is the kinematically distinct nature of the northern h i complex .
as demonstrated in paper i , this region is static along the line of sight with respect to the main body of ddo165 .
the northern complex contains @xmath015% of the total h i mass in the system ; column densities exceed 10@xmath12 @xmath13 and young stars are associated with this gas cloud . using h i data alone , both an infall and an outflow origin for this kinematic discontinuity remain viable .
the energies available from our sfh analysis , as well as the chemical homogeneity found by @xcite , favor a scenario where the northern complex has a feedback - triggered outflow origin , with in situ sf proceeding in the ejected gas . stated equivalently , the @xmath010@xmath54 erg of feedback energy produced over the last 500 myr by the ongoing starburst is energetically capable of creating the giant h i hole , ejecting the northern gas cloud away from the center of the system , and inducing incoherent but high velocity motion of a large fraction of the neutral gas .
the giant h i hole in ddo165 is similar to the ones found in the m81 group dwarfs m81dwa and hoi , studied in detail by @xcite . as that work and
the present results indicate , feedback from recent sf is energetically capable of producing each of these kpc - size structures .
ddo165 and hoi have neutral gas masses in the 10@xmath65 m@xmath18 range ; a characteristic cumulative feedback energy of 10@xmath66 deposited on 100 myr timescales is capable of creating kpc - size h i holes .
the neutral gas mass and cumulative feedback energy in m81dwa are each an order of magnitude lower over the same characteristic time interval .
while our sfh analysis allows us to conclude that sf can create such giant h i structures , we are left puzzled as to why it does not do so in other m81 group dwarf galaxies with similar normalized sfrs and neutral gas masses .
the creation of giant h i structures appears to be efficient in galaxies less massive than @xmath010@xmath63 m@xmath18 that host temporally extended sf events that include a significant fraction ( perhaps 30% or more ) of all stars in the system .
ddo165 is an extreme example of how feedback drives the evolution of dwarf galaxies .
the gyr - long burst clearly indicates that low - mass galaxies are capable of vigorous sf over timescales much longer than the canonical instantaneous burst .
the resulting energy outputs from such long sf episodes are sufficient to induce significant bulk motions in the surrounding ism .
the signatures of feedback in ddo165 are diverse , including coherent and incoherent gas motions .
ddo165 serves as a poignant reminder that a remarkably active past can be unlocked by probing the sfhs of nearby galaxies .
we thank an anonymous referee for an especially detailed , thorough and insightful report that helped to significantly strengthen this manuscript .
support for this work was provided by nasa through grant go-10605 from the space telescope science institute , which is operated by aura , inc .
, under nasa contract nas5 - 26555 .
j.m.c . and h.p.m .
thank macalester college for research support .
d.r.w . and s.r.w .
are grateful for support from penrose fellowships .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration , and nasa s astrophysics data system .
bastian , n. , weisz , d.r . ,
skillman , e.d . ,
mcquinn , k.b.w . , dolphin , a.e . , gutermuth , r.a . , cannon , j. m. , ercolano , b. , gieles , m. , kennicutt , r. c. , lee , j. c. , & walter , f. 2011 , , in press ( arxiv/1010.1837 ) warren , s. r. , weisz , d. r. , skillman , e. d. , cannon , j. m. , dalcanton , j. j , dolphin , a. e. , kennicutt , r. c. , koribalski , b. , ott , j. , stilp , a. m. , van dyk , s. d. , walter , f. , & west , a. a. 2011 , , submitted lccccccc 1 & 13:06:17.1 & 67:43:04.4 & 975@xmath1325 , 40&281 & 1.16 & - & - + 2 & 13:06:22.1 & 67:41:46.9 & 520 & 260 & 1.05 & 11 & 23.0 + 4 & 13:06:27.7 & 67:41:04.5 & 390 & 195 & 1.96 & - & - + 5 & 13:06:30.1 & 67:41:26.5 & 737 & 369 & 1.59 & 7 & 51.2 + 6 & 13:06:30.6 & 67:42:31.0 & 2210@xmath11084 , 130&774 & 0.73 & - & - + 7 & 13:06:38.6 & 67:41:33.7 & 455 & 228 & 2.14 & 9 & 24.6 + [ t1 ] lcccc h i hole 1 & 13:06:17.1 & 67:43:04.4 & 45@xmath115 , 40&2888 + h
i hole 2 & 13:06:22.1 & 67:41:46.9 & 24 & 2690 + h i hole 4 & 13:06:27.7 & 67:41:04.5 & 18 & 310 + h
i hole 5 & 13:06:30.1 & 67:41:26.5 & 34 & 2126 + h i hole 6 & 13:06:30.6 & 67:42:31.0 & 102@xmath150 , 310&28439 + h i hole 7 & 13:06:38.6 & 67:41:33.7 & 21 & 701 + cluster1 ( cl1 ) & 13:06:14.6 & 67:42:23.5 & 20 & 2307 + cluster2 ( cl2 ) & 13:06:17.8 & 67:42:13.5 & 20 & 2321 + northern1 ( n1 ) & 13:06:24.4 & 67:43:41.2 & 12 & 261 + northern2 ( n2 ) & 13:06:24.4 & 67:43:41.2 & 24 & 957 + control region1 ( c1 ) & 13:06:11.1 & 67:42:32.9 & 20 & 1901 + control region2 ( c2 ) & 13:06:12.1 & 67:43:03.4 & 20 & 1274 + control region3 ( c3 ) & 13:06:12.4 & 67:42:02.8 & 20 & 1631 + control region4 ( c4 ) & 13:06:16.6 & 67:43:38.0 & 20 & 615 + control region5 ( c5 ) & 13:06:17.0 & 67:41:48.4 & 20 & 1573 + control region6 ( c6 ) & 13:06:17.1 & 67:41:23.7 & 20 & 2224 + control region7 ( c7 ) & 13:06:17.4 & 67:42:37.4 & 20 & 681 + control region8 ( c8 ) & 13:06:21.0 & 67:42:47.9 & 20 & 2056 + control region9 ( c9 ) & 13:06:21.6 & 67:42:24.5 & 20 & 2566 + control region10 ( c10 ) & 13:06:23.1 & 67:43:17.4 & 20 & 1366 + control region11 ( c11 ) & 13:06:24.5 & 67:41:20.8 & 20 & 730 + control region12 ( c12 ) & 13:06:24.6 & 67:42:08.5 & 20 & 2410 + control region13 ( c13 ) & 13:06:27.9 & 67:41:54.1 & 20 & 2368 + control region14 ( c14 ) & 13:06:29.8 & 67:43:29.1 & 20 & 684 + control region15 ( c15 ) & 13:06:34.3 & 67:43:16.2 & 20 & 1008 + control region16 ( c16 ) & 13:06:35.1 & 67:41:22.5 & 20 & 534 + control region17 ( c17 ) & 13:06:37.4 & 67:42:52.4 & 20 & 1539 + control region18 ( c18 ) & 13:06:40.2 & 67:41:58.4 & 20 & 1300 + control region19 ( c19 ) & 13:06:41.0 & 67:42:27.5 & 20 & 1354 + control region20 ( c20 ) & 13:06:42.1 & 67:43:07.8 & 20 & 718 + [ t2 ] lcccc c1 & 13:06:10.2 & 67:41:42.2 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c2 & 13:06:13.9 & 67:42:22.5 & ( [email protected])@xmath110@xmath63 & ( [email protected])@xmath110@xmath67 + c3 & 13:06:14.6 & 67:41:30.5 & ( [email protected])@xmath110@xmath63 & ( [email protected])@xmath110@xmath67 + c4 & 13:06:14.8 & 67:43:13.4 & no fit & no fit + c5 & 13:06:15.4 & 67:42:20.1 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c6 & 13:06:16.3 & 67:42:05.8 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c7 & 13:06:17.9 & 67:42:13.9 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c8 & 13:06:18.1 & 67:41:34.7 & no fit & no fit + c9 & 13:06:20.0 & 67:43:49.4 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath68 + c10 & 13:06:22.6 & 67:42:02.3 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c11 & 13:06:23.7 & 67:42:37.9 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c12 & 13:06:24.0 & 67:42:58.5 & ( [email protected])@xmath110@xmath63 & ( [email protected])@xmath110@xmath67 + c13 & 13:06:25.8 & 67:42:31.2 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c14 & 13:06:29.1 & 67:42:27.2 & ( [email protected])@xmath110@xmath63 & ( [email protected])@xmath110@xmath69 + c15 & 13:06:29.3 & 67:41:28.7 & ( [email protected])@xmath110@xmath63 & ( [email protected])@xmath110@xmath67 + c16 & 13:06:29.8 & 67:42:14.8 & ( [email protected])@xmath110@xmath61 & ( [email protected])@xmath110@xmath68 + c17 & 13:06:30.2 & 67:41:52.0 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c18 & 13:06:30.8 & 67:42:17.2 & ( [email protected])@xmath110@xmath63 & ( [email protected])@xmath110@xmath69 + c19 & 13:06:30.8 & 67:42:41.2 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + c20 & 13:06:35.8 & 67:42:07.1 & ( [email protected])@xmath110@xmath65 & ( [email protected])@xmath110@xmath67 + [ t3 ] lccccccccc + + 1 & 7 & 0.1 & 2.7 & 0.094 & 2.5 & 0.090 & 2.4 & 0.090 & 2.4 + 4 & 7 & 0.1 & 0.86 & 0.059 & 0.47 & 0.062 & 0.50 & 0.057 & 0.46 + 6 & 7 & 0.1 & 63 & 0.081 & 50 & 0.094 & 59 & 0.11 & 67 + + + + 1 & 11.2 & 0.1 & 5.2 & 0.094 & 4.8 & 0.090 & 4.6 & 0.090 & 4.6 + 4 & 12.0 & 0.1 & 1.8 & 0.059 & 1.0 & 0.062 & 1.1 & 0.057 & 0.97 + 6 & 12.0 & 0.1 & 130 & 0.081 & 110 & 0.094 & 130 & 0.11 & 140 + + + + 1 & 8.5 & 0.1 & 3.5 & 0.094 & 3.3 & 0.090 & 3.1 & 0.090 & 3.1 + 4 & 8.5 & 0.1 & 1.1 & 0.059 & 0.62 & 0.062 & 0.66 & 0.057 & 0.60 + 6 & 8.5 & 0.1 & 83 & 0.081 & 65 & 0.094 & 77 & 0.11 & 87 + [ t5 ] lccccc 1 & 32 & 3.1 & 8.4 & 37% & @xmath2430 myr + 2 & 23 & 3.7 & 28 & 13% & @xmath2415 myr + 4 & 22 & 0.06 & n / a & n / a & @xmath0330 myr + 5 & 52 & 4.3 & 76 & 5.7% & @xmath2425 myr + 6 & 89 & 87.4 & 6170 & 1.4% & @xmath2420 myr + 7 & 25 & 0.90 & n / a & n / a & @xmath0130 myr + [ t6 ] lccc h i hole 1 & 2.5 & 24 & 110 + h i hole 2 & 28 & 270 & 1400 + h i hole 4 & 0.0 & 0.0 & 0.0 + h i hole 5 & 6.8 & 60 & 230 + h i hole 6 & 210 & 1700 & 7500 + h i hole 7 & 0.0 & 0.0 & 0.4 + cluster1 ( cl1 ) & 33 & 290 & 980 + cluster2 ( cl2 ) & 280 & 1100 & 2400 + northern1 ( n1 ) & 15 & 75 & 180 + northern2 ( n2 ) & 34 & 130 & 270 + control region1 ( c1 ) & 6.5 & 58 & 240 + control region2 ( c2 ) & 1.5 & 13 & 47 + control region3 ( c3 ) & 4.9 & 44 & 170 + control region4 ( c4 ) & 0.9 & 8.1 & 29 + control region5 ( c5 ) & 21 & 51 & 160 + control region6 ( c6 ) & 0.4 & 3.3 & 9.6 + control region7 ( c7 ) & 15 & 130 & 470 + control region8 ( c8 ) & 3.0 & 29 & 160 + control region9 ( c9 ) & 17 & 150 & 500 + control region10 ( c10 ) & 42 & 110 & 190 + control region11 ( c11 ) & 0.0 & 0.0 & 0.0 + control region12 ( c12 ) & 70 & 380 & 1300 + control region13 ( c13 ) & 20 & 210 & 1100 + control region14 ( c14 ) & 2.0 & 17 & 48 + control region15 ( c15 ) & 0.0 & 0.4 & 9.7 + control region16 ( c16 ) & 0.0 & 0.1 & 2.5 + control region17 ( c17 ) & 7.7 & 67 & 250 + control region18 ( c18 ) & 49 & 190 & 510 + control region19 ( c19 ) & 7.5 & 69 & 320 + control region20 ( c20 ) & 0.0 & 0.2 & 5.0 + total galaxy & 1900 & 11000 & 35000 + [ t7 ] | we compare the stellar populations and complex neutral gas dynamics of the m81 group dirr galaxy ddo165 using data from the _ hst _ and the _ vla_. paper i identified two kinematically distinct h i components , multiple localized high velocity gas features , and eight h i holes and shells ( the largest of which spans @[email protected] kpc ) .
using the spatial and temporal information from the stellar populations in ddo165 , we compare the patterns of star formation over the past 500 myr with the h i dynamics .
we extract localized star formation histories within 6 of the 8 h i holes identified in paper i , as well as 23 other regions that sample a range of stellar densities and neutral gas properties . from population synthesis modeling , we derive the energy outputs ( from stellar winds and supernovae ) of the stellar populations within these regions over the last 100 myr , and compare with refined estimates of the energies required to create the h i holes . in all cases , we find that `` feedback '' is energetically capable of creating the observed structures in the ism
. numerous regions with significant energy inputs from feedback lack coherent h i structures but show prominent localized high velocity gas features ; this feedback signature is a natural product of temporally and spatially distributed star formation . in ddo165 , the extended period of heightened star formation activity ( lasting more than 1 gyr ) is energetically capable of creating the observed holes and high velocity gas features in the neutral ism . |
why are precision tests of solar models important ? the sun is a laboratory in which the predictions of stellar evolution theory can be compared with uniquely accurate and detailed measurements .
stellar evolution calculations are used throughout astronomy to classify , date , and interpret the spectra of individual stars and of galaxies .
comparisons , discussed in this paper , between helioseismological measurements and solar model calculations suggest that at least one of the ingredients of stellar evolution calculations is not known as precisely as previously believed .
we shall see that there are reasons for questioning the accuracy of the most sophisticated and detailed determinations of stellar abundances , the recent measurements of the solar heavy element abundances .
alternatively , unexpectedly large changes could be required in the radiative opacity .
however , we shall also see that the disagreement between helioseismological measurements and solar model predictions ( with the new metal abundances ) occur at relatively low temperatures and therefore do not affect significantly the predicted solar neutrino fluxes .
helioseismology provides sensitive and powerful tests of the theory of stellar evolution .
in addition to measuring the depth of the solar surface convection zone and the surface helium abundance , inversions of seismic data are used to measure to high precision the speed of sound as a function of depth in the star for almost the entire solar interior .
the density distribution can also be determined , although with an order of magnitude less precision than for the sound speed .
a number of investigators have made comparisons of seismic data with solar models and have confirmed that the standard solar mixture of grevesse & noels ( 1993 ) and the updated mixture of grevesse & sauval ( 1998 ) yield solar models in good agreement with the data ( e.g. , bahcall & pinsonneault 1995 ; christensen - dalsgaard , et al .
1996 ; bahcall , pinsonneault & basu 2001 ; christen - dalsgaard 2002 ; couvidat , turck - chieze , & kosovichev 2003 ; sackmann & boothroyd 2003 ; richard , thado , & vauclair 2004 , and references therein . )
as early as 1988 , bahcall & ulrich ( 1988 ) showed that detailed solar models computed with the accurate physics and the numerical precision required for solar neutrino predictions yielded results in good agreement with the then - available helioseismological data . in a series of papers that preceded the epochal and definitive measurements of the sno and super - kamiokande solar neutrino experiments ( ahmad , et al . 2001 ; fukuda et al . 2001 ) , we showed that the excellent agreement between the computed sound speeds in precise standard solar models and the precise helioseismological inversions ( differences @xmath4% rms throughout the solar interior)implied that new physics was required to solve the solar neutrino problem ( bahcall , pinsonneault , basu , & christensen - dalsgaard 1997 ; bahcall , basu , & pinsonneault 1998 ; bahcall , pinsonneault , & basu 2001 ; bahcall 2001 ) .
new and more powerful analyses of the surface chemical composition of the sun have recently become available .
these new analyses use three - dimensional atmospheric models , take account of hydrodynamic effects , and pay special attention to uncertainties in the atomic data and the observed spectra .
lower mass fractions have been obtained in this way for c , n , o , ne , and ar ( asplund et al . 2000 ; asplund 2000 ; allende prieto , lambert , & asplund 2001 ; allende prieto , lambert , & asplund 2002 ; asplund et al .
the new abundance determinations , together with the previous best - estimates for other solar surface abundances ( grevesse & sauval 1998 ) , imply @xmath5 , much less than the previous value of @xmath6 ( grevesse & sauval 1998 ) .
in fact , the recent estimates for the c , n , and o mass fractions are lower than all the abundance measurements we have used in the precision solar models in this series going back to 1971 ( see , e.g. , table ii of bahcall and & pinsonneault 1995 ) . despite the great improvement in the techniques
now used to determine the new element abundances , the new abundances cause the depth calculated for the solar convective zone with the aid of a standard solar model , @xmath7 ( bahcall & pinsonneault 2004 ; bahcall , serenelli , & pinsonneault 2004 ; basu & antia 2004 ) , to be in strong disagreement with the measured depth , @xmath8 which is determined by helioseismological techniques ( kosovichev & fedorova 1991 ; christensen - dalsgaard , gough , & thompson 1991 ; guzik & cox 1993 ; basu & antia 1997 , 2004 ; basu 1998 ) .
paradoxically , the calculated depth of the convective zone obtained using the older element abundances , @xmath9 , agrees with the helioseismological value ( bahcall , pinsonneault , & basu 2001 ) .
this situation has been described as the `` convective zone problem '' ( bahcall , serenelli , & pinsonneault 2004 ) .
our goal here is to determine the helioseismological implications of the recent abundance determinations .
we compare the helioseismologically measured depth of the solar convective zone , the sound speed and density as a function of radius , and the primordial helium abundance with the values that are obtained using a series of precise solar models .
the solar models considered here incorporate the most recent and accurate nuclear and stellar data , including the equation of state and radiative opacity . we describe in [ sec : models ] the solar models whose properties are investigated in the present paper .
we then discuss in
[ sec : inversions ] the helioseismological data and the inversion technique that we have used to obtain the measured depth of the convective zone , the sound speeds and density as a function of radius , and the initial helium abundance .
we compare in
[ sec : comparison ] the properties of our set of solar models with the solar parameters that are determined by helioseismology .
we summarize and discuss our conclusions in [ sec : conclusions ] .
we describe in this section the basic ingredients of six solar models that we use to assess the helioseismological implications of the recent heavy element abundance determinations .
the six solar models considered in detail in this paper are listed below .
models 1 and 2 were originally computed by bahcall & pinsonneault ( 2004 ) ; model 4 was computed by bahcall , pinsonneault , serenelli ( 2004 ) .
( 1 ) bp04 : : : older element abundances from grevesse , & sauvall ( 1998 ) , and best - available values for all other input parameters ( including improved nuclear rates and equation of state ) ; ( 2 ) bp04 + : : : the same as bp04 except that recent lower estimates for heavy element abundances are incorporated ; ( 3 ) bp04eos96 : : : the same as bp04 but with the opal 1996 eos ( rogers , swenson , & iglesias 1996 ) instead of the opal 2001 eos ( rogers 2001 , rogers and nayfonov 2002@xcite ) ; ( 4 ) bp04 + 21% : : : the same as bp04 + except that the opal radiative opacity is increased by 21% near the base of the convective zone ; ( 5 ) bp04 + 11% : : : the same as bp04 + except that the opal radiative opacity is increased by 11% for temperatures ranging from @xmath10k to @xmath2k ; and ( 6 ) bp00 : : : our best previous - generation standard solar model , obtained by bahcall , pinsonneault , & basu ( 2001 ) with older values of nuclear reaction data , an older equation of state ( opal 1996 ) , and the grevesse & sauval ( 1998 ) element abundances .
the code and techniques used in these calculations have been described in bahcall & pinsonneault ( 1992 , 1995 ) , bahcall & ulrich ( 1988 ) , bahcall , pinsonneault , & basu ( 2001 ) .
the reader may wonder why we include in this paper the results from the model bp00 , when bp04 has superseded bp00 by incorporating more accurate nuclear reaction data and an improved equation of state .
we include results from bp00 as well as bp04 in order to have some indication of the kind of differences that can be expected , independent of solar abundance determinations , as further improvements are made in the input data to solar models .
the differences between values obtained with the bp00 and the bp04 models may be regarded as within the expected range .
we shall see in what follows that the differences in solar model results caused by adopting the new heavy element abundance determinations are much larger than the differences between the results obtained with bp00 and bp04 .
we want our investigations to be as precise as possible and our inferences to be as free as possible from dependence upon the idiosyncrasies of a particular stellar evolution code .
therefore , we have recalculated the bp04 , bp04 + and bp04 + 21% solar models using the garching stellar evolution code ( see , e.g. , schlattl , weiss , & ludwig 1997 and schlattl 2002 for details of the code ) , to which the nuclear energy generation routine ` exportenergy.f ' has been coupled
. the nuclear cross sections adopted are those used in bahcall & pinsonneault ( 2004 ) .
the models were calculated using the latest version of the opal equation of state ( rogers 2001 ) , opal radiative opacities ( see below for the composition adopted ) and element diffusion for helium and metals ( thoul , bahcall , & loeb 1994 , code available at the url given in footnote 1 ) .
the mixing length theory for convection has been used in all the models .
the schwarzschild criterion was used to determine the location of the convective boundaries .
we have verified that the garching stellar evolution code and the bahcall - pinsonneault code ( which has its origins in the caltech , ucla , and yale codes , see bahcall & ulrich 1988 ; bahcall & pinsonneault 1992 , 1995 ; prather 1976 ; pinsonneault , kawaler , sofia , & demarque 1989 ) ) yield identical results to the accuracy of interest in all of the investigations considered in this paper . the model , bp04 , which was calculated assuming the older grevesse & sauval ( 1998 ) solar surface composition , has a present surface ratio of heavy element to hydrogen mass fractions of @xmath11 .
the model bp04 + , which incorporates the new determinations of the solar heavy element composition , has a much lower ratio of heavy elements to hydrogen , @xmath12 .
since new solar abundance determinations are being reported as they come available , table 1 of bahcall , serenelli , & pinsonneault ( 2004 ) lists the specific element abundances adopted in computing both bp04 and bp04 + .
the model bp04 + 21% was designed to bring into agreement the calculated and the helioseismologically measured depths of the convective zone using a solar model that incorporates the recent heavy element abundance determinations .
bahcall , serenelli , & pinsonneault ( 2004 ) showed that a local 21% increase in the tabulated opal radiative opacity near the base of the convective envelope will produce a model with the base of its convective zone at @xmath13 , in essentially perfect agreement with the measured value for the depth of the convective zone .
the factor by which the opacity was increased is similar to the factor needed by basu & antia ( 2004 ) to construct solar envelope models with the new heavy element abundances that have the same convection zone depth , helium abundance , and density profile as the sun .
very recently , seaton & badnell ( 2004 ) have shown that a detailed calculation using the methods of the opacity project ( op , see seaton et al .
1994 ) for a six element mixture yields a rosseland - mean opacity in the region of interest of order 5% larger than the opal opacity for the same mixture .
all the models assume a solar age of @xmath14 yr , a present solar luminosity @xmath15 ergs@xmath16 , and a present solar radius of r@xmath17 cm .
for each model , opal opacity tables were used that correspond to the detailed composition that was adopted .
we summarize in this section the largely standard techniques that we use to determine the differences between the solar model characteristics and the properties of the sun as determined by helioseismological measurements .
helioseismological inversions generally proceed through a linearization of the equations of stellar oscillation , using their variational property , around a known reference model ( see , e.g. , dziembowski , pamyatnykh & sienkiewicz 1990 ; dppen et al .
1991 ; antia & basu 1994 ; dziembowski et al . 1994
; elliott 1995 ; tripathy and christensen - dalsgaard 1998 ) . the differences between the structure of the sun and the reference model are then related to the differences in the measured oscillation frequencies of the sun and the model by known kernels . non - adiabatic effects and other errors in modeling the surface layers give rise to frequency shifts which are not accounted for by the variational principle ( cox & kidman 1984 ; balmforth 1992 ) .
since the eigenfunctions of low- and medium - degree modes are essentially independent of degree in the near - surface layers , the frequency shifts are just a function of mode frequency , divided by the mode inertia ( christensen - dalsgaard & berthomieu 1991 ; christensen - dalsgaard & thompson 1997 ) .
the frequency of a deeply penetrating mode is shifted less by near - surface perturbations than that of a shallowly penetrating mode of the same frequency . in the absence of any first - principle formulation ,
these effects are usually taken into account in an _ ad hoc _ manner by including an arbitrary function of frequency in the variational formulation ( dziembowski et al .
thus , the fractional change in the frequency of a mode can be expressed in terms of the fractional changes in the structure of the model , which can be characterized , for example , by the adiabatic sound speed , @xmath18 , and the density , @xmath19 , as well as a surface term .
after linearization , one obtains : @xmath20 ( e.g. , dziembowski et al .
1990 ) . here
@xmath21 is the difference in the frequency @xmath22 of the @xmath23th mode between the solar data and a reference model , @xmath23 representing the pair @xmath24 , where @xmath25 is the radial order and @xmath26 the degree of the model .
the kernels @xmath27 and @xmath28 are known functions of the reference model which relate the changes in frequency to the changes in @xmath29 and @xmath19 , respectively .
the term involving @xmath30 takes into account the near - surface errors in modeling the structure and the modes , and @xmath31 is the mode inertia of the @xmath32 mode .
equation ( [ eq : inv ] ) constitutes the inverse problem that must be solved to infer the differences in structure between the sun and the reference model .
the inversions shown in this paper have been carried out using the the subtractive optimally localized averages ( sola ) technique ( pijpers & thompson 1992 , 1994 ) .
details of how sola inversions are carried out and how various parameters of the inversion are selected are given by rabello - soares , basu , & christensen - dalsgaard ( 1999 ) . in this paper
, we use helioseismic inversions to determine how similar the different solar models discussed in [ sec : models ] are to the real sun .
each of the models described in
[ sec : models ] is used as a reference model . for the helioseismological data , we use solar oscillation frequencies obtained by the michelson doppler imager ( mdi ) on board the solar and heliospheric observatory ( soho ) . in particular
, we use frequencies obtained from mdi data that were collected for the first 360 days of its observation ( schou et al .
this data set was chosen because it was derived from a long time series when solar activity was low .
the length of the time series results in reduced noise , and hence a larger number of modes for which the frequencies can be determined reliably .
mode - sets derived from longer data sets are available , but they only consist of low degree modes ( e.g. , bertello et al . 2000 ) .
also , a longer time series would have meant adding observations from periods of increasing solar activity , which would have changed the frequencies .
it is a well established fact that solar frequencies increase with solar activity .
however , it is also known that the increase occurs as an increase in the surface term in eq .
[ eq : inv ] , and hence does not change inversion results ( basu 2002 ) .
we invert for both the sound - speed differences and the density differences between the solar models and the sun .
in this section , we compare solar parameters determined from helioseismological measurements with the values obtained from the six solar models that are discussed in [ sec : models ]
. table [ tab : comparisons ] summarizes the principal results .
figure [ fig : velocitydiffs ] shows the fractional differences between the sound speeds as a function of solar radius that are computed for each of the solar models and the sound speeds determined from helioseismology .
figure [ fig : densitydiffs ] shows for the density profiles a similar trend as figure [ fig : velocitydiffs ] shows for the sound speeds . since it is well known that sound speed determinations are more accurate and more robust than density determinations , we do not discuss further the density profiles other than to remark that they are consistent with all of the other comparisons we make between solar model predictions and helioseismological measurements .
.solar model predictions versus helioseismological determinations .
the table presents for comparison with helioseismological measurements the results of a series of four solar models discussed in [ sec : models ] .
the successive columns give the model designation , the adopted present heavy element to hydrogen mass ratio at the solar surface , the rms fractional difference between the solar model sound speeds and the helioseismologically - determined sound speeds , the rms fractional difference for the density , the radius of the convective zone , and the present surface helium abundance . for consistency , all the results reported in this table were obtained with the garching stellar evolution code .
[ tab : comparisons ] [ cols="<,^,^,^,^,^",options="header " , ] the third column of table [ tab : comparisons ] presents the fractional rms differences between each solar model ( used as a reference model , see
[ sec : inversions ] ) and the helioseismologically determined sound speeds .
we see that the bp00 and the bp04 solar models , both of which are computed using the older grevesse & sauval ( 1998 ) heavy element abundances , are in good agreement with the solar sound speeds .
the rms agreement with the solar sound speeds is about 0.1% for both bp00 and bp04 .
figure [ fig : velocitydiffs ] shows the agreement between the sound speeds predicted by the bp00 solar model ( dark line ) and the bp04 solar model ( dashed line ) .
the fifth column of table [ tab : comparisons ] shows that the calculated depth of the convective zone for the bp00 and the bp04 models is in satisfactory agreement with the the measured value of the depth of the convective zone given in equation([eq : radiuscz ] ) , @xmath33 .
the surface helium abundance of the sun has recently been redetermined by basu and antia ( 2004 ) .
they find @xmath34 the interpretation of the error given in equation ( [ eq : yhelio ] ) is not simple since systematic uncertainties are dominant .
the fifth column of table [ tab : comparisons ] shows that the present - day surface helium abundance obtained from models bp00 and bp04 may be slightly lower than is obtained from helioseismology , but the statistical significance of this difference is uncertain . for completeness , we have computed a model that is identical to bp04 except that instead of using the 2001 opal equation of state , as was done in deriving the model bp04 , we use the older 1996 opal equation of state .
the results are given in the third row of table [ tab : comparisons ] .
improvements in the equation of state between 1996 and 2001 are reflected in table [ tab : comparisons ] by the slightly different values that are found for bp04 ( row two ) and bp04eos96 ( row three ) .
we see that the improvement in the equation of state does not affect significantly the agreement of the solar model results with the measured helioseismological values .
we conclude that plausible changes in the equation of state are unlikely to explain the discrepancy between solar model predictions and helioseismological measurements when the lower metal abundances are used .
as discussed in basu , pinsonneault , and bahcall ( 2001 ) , the effect of mixing in the radiative zone of the sun would be in the direction to reconcile the meteoritic and solar photospheric lithium abundances and to bring the computed surface helium slightly closer to the measured value .
such models have a somewhat shallower solar surface convection zone and the overall agreement with the sound speed data is comparable , or slightly less good , than models without extra mixing .
figure [ fig : velocitydiffs ] shows the dramatic lack of agreement between the helioseismological sound speeds and the values predicted by the bp04 + solar model , which uses the new heavy element abundance determinations that lead to @xmath5 .
the biggest discrepancy is in the vicinity of the base of the convective zone , near @xmath35 .
however , there is a significant discrepancy between bp04 + and the helioseismological values all the way into about @xmath36 .
table [ tab : comparisons ] summarizes the magnitude of this discrepancy . for the solar model bp04 + , the rms discrepancy in the sound speeds is more than a factor of three worse than for the bp04 model ( and more than a factor of five worse than for the bp00 model ) .
furthermore , the depth of the convective zone , @xmath37 , given in column 6 of table [ tab : comparisons ] is inconsistent with the measured value of @xmath38.finally , the surface helium abundance given in column 6 , @xmath39 , is lower than the measured value given in equation ( [ eq : yhelio ] ) .
we conclude that the solar model bp04 + , which is constructed using the most recent heavy element abundance estimates , is inconsistent with helioseismological measurements .
the comparison between the predictions of the model bp04 + 21% and the helioseismological data is very instructive .
this solar model was investigated in bahcall , serenelli , & pinsonneault ( 2004 ) because the 21% increase in the radiative opacity relative to the standard opal opacity was found to be sufficient to resolve the discrepancy in the calculated depth of the convective zone that was obtained with bp04 + model ( with no enhanced opacity ) . for a related discussion , see the paper by tripathy , basu , and christensen - dalsgaard 1998 .
the bp04 + 21% model was constructed with exactly the same input data as for the bp04 + model , including the recent heavy element abundance determinations , but in addition bp04 + 21% has the radiative opacity increased artificially by 21% near the base of the convective zone
. the precise form of the opacity increase was postulated to be of the form obtained by multiplying the opal opacity in the vicinity of the convective envelope boundary by a lorentzian function @xmath40 .
specifically , the multiplicative factor @xmath40 was taken to be @xmath41 here @xmath42 is the temperature in the solar model .
the perturbed opacity is @xmath43 , where @xmath44 is the unperturbed radiative opacity , @xmath45 is the amplitude of the perturbation , and @xmath46 is the width of the perturbation ( defined as the point where the perturbation drops to @xmath47 ) .
the temperature at the base of the cz is @xmath48k , which was used for @xmath49 in equation ( [ eq : defnofperturbation ] ) .
the bp04 + 21% solar model was calculated for a width of the opacity perturbation @xmath50 .
this value of @xmath46 corresponds to a width in the solar radius of only @xmath51 .
figure [ fig : velocitydiffs ] shows two things about the solar model .
first , the 21% increase in the opacity near the base of the solar convective zone indeed improves significantly the agreement with the measured sound speeds over what is obtained with the model bp04 + .
second , the improved agreement is limited to the region near the base of the convective zone and there remains a significant disagreement down to radii of order @xmath52 ( @xmath53k ) .
of course , different assumed forms of the factor @xmath40 lead to different estimates of how much opacity change is required to construct a model with the correct depth of the convection zone ( see , e.g. , basu & antia 2004 ) . in summary ,
figure [ fig : velocitydiffs ] indicates that the radiative opacity would have to be changed in a broad range of temperatures ( radii ) in order to resolve the discrepancies between helioseismological measurements and solar model predictions made using the new heavy element abundances .
the relatively low value for the surface helium abundance , @xmath54 obtained with ( see table [ tab : comparisons ] ) , may also reflect the need for an opacity correction that extends down to @xmath55k .
motivated by the results of
[ subsec : bp04 + 21 ] , we have computed a variety of solar models assuming the correctness of the recently determined low metal abundances but with different assumed opacity changes .
we have studied the helioseismological properties of these models .
the reader will immediately recognize that one can in principle consider an infinite number of such ` low - metal , higher - opacity ' models , with prescriptions for changing the opacity of varying complexity and artificiality .
we acknowledge that there is limited utility in computing such models without a physical basis for the assumed opacity changes .
however , we have found a relatively simple prescription for changing the opacity , while adopting the low metal abundances , that yields reasonable agreement with the observed helioseismological properties .
we present the results for this model here not out of any conviction that the assumed opacity law is correct , but rather to illustrate the general quality of the fit to the helioseismological data that is possible and to indicate approximately how much the opacity would have to be shifted in order to obtain reasonably good agreement with the helioseismological measurements .
the results of [ subsec : bp04 + 21 ] indicate that the opacity must be changed over a relatively broad range of temperatures if we adopt the lower metal abundances . for simplicity , we assumed a constant 11% increase above the opal opacity from @xmath10k ( @xmath1 ) down to @xmath2k ( @xmath56 ) , where the opacity increase was smoothly turned off ( half - width of turn off is @xmath57k ) .
we denote this model by bp04 + 11% .
we are sure that the prescription of a constant opacity increase that is implemented in bp04 + 11% is too simple to represent the improvements in the radiative opacity that are likely to result from detailed quantum mechanical calculations of the solar mixture of hydrogen , helium , and heavy elements .
but , we shall see that this model with a constant opacity increase fits the data reasonably well and is a crude approximation to what might guess is required by comparing ( see figure [ fig : opacitydifferences ] below ) the opacities in the bp04 model ( successful in describing the helioseismological data ) and the bp04 + model ( unsuccessful in describing the helioseismological data ) .
figure [ fig : diffsbp04 + 11 ] and table [ tab : comparisons ] show that the bp04 + 11% solar model fits the helioseismological data with an accuracy that is comparable to our best - fitting solar models , bp00 and bp04 .
we conclude that an increase in the opacity of the order of 10% in the range @xmath10k to @xmath2k would resolve the discrepancy between the predictions of solar models computed with the new lower metal abundances and the helioseismological measurements .
figure [ fig : opacitydifferences ] shows how the opal radiative opacity changes due to the adoption of the new solar composition.we have evaluated the opacity at the same temperatures and densities for two solar models , bp04 and bp04 + , that differ only in their assumed composition . to correct for the small effect that at the same temperature the density differs slightly in the two models
, we use the following equation : @xmath58 where @xmath59 here @xmath60 and @xmath61 are the opacities corresponding to the bp04 and bp04 + solar models respectively , @xmath42 and @xmath19 are temperatures and densities at a given point in the bp04 model and @xmath62 is the density in the bp04 + model at the temperature @xmath42 . including the density dependence ,
makes very little difference near the base of the cz but increases the fractional opacity difference by about 40% of its value at the highest temperature ( @xmath63 k ) at which an opacity perturbation was introduced into the solar model bp04 + 11% .
the fractional difference is small ( less than 3% ) in the regions where solar neutrinos are produced ( @xmath64 and @xmath65 k ) .
however , for @xmath66 k the difference increases and reaches about 15% close to the base of the convective zone .
figure [ fig : opacitydifferences ] shows why the bp04 + 11% model approximately restores the agreement with helioseismological measurements .
we have evolved solar models with larger opacity increases near the base of the convective zone and smaller increases further in , but we have not obtained by this procedure a substantial improvement in the agreement with helioseismological data over what is achieved with the bp04 + 11% model .
basu & antia ( 2004 ) discussed the characteristics of a solar envelope model which invoked , in order to satisfy helioseismic constraints , a 19% increase in radiative opacity relative to the tabulated opal opacities .
we have evolved a full solar model with the same opacity increase and heavy element abundance as the basu & antia ( 2004 ) model ( i.e. a 19% increase in opacity from the base of the convection zone to a temperature of @xmath67k and the heavy element to hydrogen mass ratio of z / x=0.0171 ) .
we find , as expected from the previous discussion of bp04 + 11% and from figure [ fig : diffsbp04 + 11 ] , that the 19% increase in opacity is too large to provide a good fit to the helioseismological data .
the depth of the convection zone for the evolved model is @xmath68 and the rms fractional sound speed discrepancy is @xmath69 .
the reason for the difference in our conclusion and the basu & antia ( 2004 ) result almost certainly lies in the fact that the basu & antia envelope model was forced to have abundance profiles near the base of the convective zone that are different from what we find in our stellar evolution models , while at the same time being silent about the helioseismological properties in the radiative interior .
the basu & antia envelope model was forced to have heavy - element and helium profiles in agreement with the helioseismological determinations near the base of the convective zone . for standard solar models ,
the heavy element and helium profiles are different from that of the sun near the base of the convective zone ( basu & antia 1994 ; bahcall , pinsonneault , basu , & christensen - dalsgaard 1997 ; antia & chitre 1998 ) , probably because of turbulent mixing not included in the standard models ( elliott & gough 1998 ) . over the radiative interior of the sun , @xmath70 to @xmath71 standard solar models like bp00 or bp04
are , as we have seen , in excellent agreement with the helioseismological data .
we summarize and discuss in this section our five principal conclusions .
the main quantitative results of our studies are given in table [ tab : comparisons ] and in figure [ fig : velocitydiffs ] and figure [ fig : diffsbp04 + 11 ] .
( 1 ) larger heavy element abundances yield satisfactory solar models .
: : standard solar models constructed with the older ( i. e. , higher ) heavy element abundances ( models bp00 , bp04 , and bp04-eos96 ) are in good agreement with the helioseismological data .
the solar sound speeds , depth of the convective zone , and surface abundance of helium determined from helioseismology are all in agreement with the values obtained from these solar models that were computed using the older element abundances .
we interpret the differences between the predictions of the models bp00 , bp04 , and bp04eos96 as indicating the expected range of characteristic parameters that can occur with typical improvements in the input data to the solar models .
( 2 ) standard models with less heavy elements disagree with helioseismology .
: : a solar model constructed with the new heavy element abundances , bp04 + , is inconsistent with the helioseismologically measured sound speeds , the depth of the convective zone , and the surface helium abundance .
( 3 ) increasing the opacity near the base of the cz helps , but is not enough . : : we calculate a solar model using the newer heavy element abundances and also increase the radiative opacity near the base of the solar convective zone by just the amount required to make the cz depth calculated with new heavy element abundances agree with the measured depth . the improved agreement between the solar model and the helioseismological determinations
is limited , like the assumed change in the radiative opacity , to regions near the base of the convective zone .
( 4 ) a 11% increase in opacity over a broader range is okay .
: : suppose that a change in the opal radiative opacity is required to explain the reason why solar models constructed with the newer heavy element abundances are in conflict with helioseismology measurements
. then the opal opacity must be increased by about 11% from about @xmath72k at the base of the cz ( @xmath73 ) all the way down to about @xmath74k ( @xmath56 ) .
it would be very useful to study whether such a change in opacities is consistent with other astronomical data .
the required 11% increase is larger than the difference reported by seaton & badnell ( 2004 ) between the radiative opacities calculated independently by the opacity project and by the opal project .
( 5 ) the predicted solar neutrino fluxes are not significantly affected .
: : the differences in the predicted solar neutrino fluxes for the most different solar models considered in this paper , bp04 and bp04 + , are all within the @xmath75 quoted theoretical errors ( see table 1 of bahcall and pinsonneault 2004 ) . if we compare models that differ only in whether or not a 11% increase in opacity has been included , the differences in predicted neutrino fluxes are slightly smaller , especially for the most important neutrino sources : 1% ( @xmath76 neutrinos ) , 2% ( @xmath77be neutrinos ) , and 6% ( @xmath78b neutrinos ) .
there are , in addition to the opacity , other sources of potential change in the solar model input data , most importantly the uncertainties in the measurements of the heavy element abundance and the uncertainties in the calculation of the heavy element diffusion coefficients .
the recent heavy element abundance determinations have quoted uncertainties of order 0.05 dex ( 12% ) ( see asplund et al .
2000 ; asplund 2000 ; allende prieto , lambert , & asplund 2001 ; allende prieto , lambert , & asplund 2002 ; asplund et al .
the heavy element diffusion coefficients are uncertain by about 15% ( see thoul , bahcall , and loeb 1994 ) . it may well be that the correct reconciliation of abundance determinations will involve modest adjustments relative to the present standard values of all of the factors mentioned above , namely , the abundances themselves , the diffusion coefficients , and the radiative opacity .
the increase of the radiative opacity by 11% obtained in this paper with the help of the model bp04 + 11% may be regarded as a plausible upper limit to the opacity correction that is required since it assumes no change in any of the other input parameters .
why have we not constructed and explored even more solar models with a variety of hypothetical changes in the radiative opacity , diffusion coefficients , and heavy element abundances ?
the reason is that for the opacity changes by themselves there is an infinity of conceivable corrections , with different amplitudes and shapes .
moreover , one can assume whatever changes one wants , within the quoted uncertainties , for the diffusion coefficients and the heavy element abundances .
improved calculations of the radiative opacity ( see , seaton and badnell 2004 for recent refinements ) will determine what , if any , significant refinements are implied by more accurate calculations .
once those calculations are available it will be appropriate to make new solar models to incorporate the newly calculated opacities .
j. n. b. is supported in part by nsf grant phy-0070928 .
s. b. was partially supported by nsf grants atm 0206130 and atm 0348837 .
a. m. s is supported in by the w. m. keck foundation through a grant to the institute for advanced study .
we are grateful to m. seaton for stimulating comments .
this work utilizes data from the solar oscillations investigation / michelson doppler imager ( soi / mdi ) on the solar and heliospheric observatory ( soho ) .
soho is a project of international cooperation between esa and nasa .
mdi is supported by nasa contract nag5 - 13261 to stanford university .
bahcall , j. n. , basu , s. , & pinsonneault , m. h. 1998 , phys .
b , 433 , 1 bahcall , j. n. , & pinsonneault m. h. 1992 , rev .
mod . phys . ,
64 , 885 bahcall , j. n. , & pinsonneault m. h. 1995 , rev .
phys . , 67 , 781 bahcall , j. n. , & pinsonneault , m. h. 2004 , phys .
lett . , 92 , 121301 bahcall , j. n. , pinsonneault , m. h. , & basu , s. 2001 , apj , 555 , 990 bertello , l. , henney , c. j. , ulrich , r. k. , et al .
2000 , , 535 , 1066 christensen - dalsgaard , j. 2002 , rev
, 74 , 1073 christensen - dalsgaard , j. , & berthomieu , g. 1991 , in solar interior and atmosphere , eds . a. n. cox , w. c. livingston , & m. matthews ( space science series ; tucson : univ . of arizona press ) , 401 christensen - dalsgaard , j. , gough , d. o. , & thompson , m. j. 1991 , apj , 378 , 413 christensen - dalsgaard , j. , & thompson , m. j. 1997 , mnras , 284 , 527 christensen - dalsgaard , j. , et al .
1996 , science , 272 , 1286 couvidat , s. , turck - chieze , s. , & kosovichev , a. g. 2003 , apj , 599 , 1434 cox , a. n. , & kidman , r. b. 1984 , in theoretical problems in stellar stability and oscillations , ed .
a. noels , & m. gabriel ( lige : institut dastrophysique ) , 259 [ [ dppen et al .
( 1991)]da91 dppen , w. , gough , d. o. , kosovichev , a. g. , & thompson , m. j. 1991 , in challenges to theories of the structure of moderate - mass stars , ed .
d. o. gough , & j. toomre ( heidelberg : springer ) . also lecture notes in physics , 388 , 111 dziembowski , w. a. , goode , p. r. , pamyatnykh , a. a. , & sienkiewicz , r. 1994 , , 432 , 417 elliott , j. r. , gough , d. o. 1998 , apj , 516 , 475 elliott , j. r. 1995 , mnras 277 , 1567 fukuda , s. , et al .
, 2001 , phys .
86 , 5651 grevesse , n. , & noels , a. 1993 , phys .
scripta , t47 , 133 grevesse , n. , & sauval , a. j. 1998 , space sci .
, 85 , 161 richard , o. , thado , s. , & vauclair , s. 2004 , solar phys . , 220 , 243 rogers , f. j. 2001 , contrib
plasma phys . ,
41 , 179 rogers , f. j. , & nayfonov , a. 2002 , apj , 576 , 1064 rogers , f. j. , swenson , f. j. , & iglesias , c. a. 1996 , apj , 456 , 902 sackmann , i .-
boothroyd , a. i. 2003 , apj , 583 , 1024 schou , j. , christensen - dalsgaard , j. , howe , r. , et al .
1998 , in structure and dynamics of the interior of the sun and sun - like stars , ed .
s. g. korzennik , & a. wilson ( noordwijk : esa ) , esa sp-418 , 2 , 845 seaton , m. j. , & badnell , n. r. 2004 , astro - ph/0404437 seaton , m. j. , yan , y. , mihalas , d. , & pradhan , a. k. 1994 , mnras , 266 , 805 tripathy , s.c .
, basu , s. , christensen - dalsgaard , j 1998 , in provost , j. , schmider , f.x .
( eds ) , poster volume ; proc iau symp .
no , 181 ; " sounding solar and stellar interiors , universite de nice , p 129 tripathy , s. c. ; christensen - dalsgaard , j. 1998 , astronomy and astrophysics,337,579 thoul , a. a. , bahcall , j. n. , & loeb , a. 1994 , apj , 421 , 828 | we show that standard solar models are in good agreement with the helioseismologically determined sound speed and density as a function of solar radius , the depth of the convective zone , and the surface helium abundance , as long as those models do not incorporate the most recent heavy element abundance determinations .
however , sophisticated new analyses of the solar atmosphere infer lower abundances of the lighter metals ( like c , n , o , ne , and ar ) than the previously widely used surface abundances .
we show that solar models that include the lower heavy element abundances disagree with the solar profiles of sound speed and density as well as the depth of the convective zone and the helium abundance . the disagreements for models with the new abundances range from factors of several to many times the quoted uncertainties in the helioseismological measurements .
the disagreements are at temperatures that are too low to affect significantly solar neutrino emission .
if errors in the calculated opal opacities are solely responsible for the disagreements , then the corrections in the opacity must extend from @xmath0k ( @xmath1 ) to @xmath2k ( @xmath3 ) , with opacity increases of order 10% . |
what are the physical conditions leading to formation of different populations of young , isolated ( single ) neutron stars , namely anomalous x - ray pulsars ( axps ) , soft gamma - ray repeaters ( sgrs ) , dim isolated neutron stars ( xdins ) , central compact objects ( ccos ) and rotating radio transients ( rrats ) ?
all these systems involve a single , young or middle - aged neutron star ( see mereghetti 2011a for a recent review of isolated neutron stars ) . despite striking differences , like energetic soft gamma bursts peculiar to sgrs and axps , there are also striking similarities , like the periods of axp / sgrs and xdins being clustered in the same range ( @xmath2 s ) . considering the estimated birth rates of these objects together with the galactic supernova rate ,
it is likely that there are evolutionary connections between some of these populations ( keane & kramer 2008 ) .
recent efforts concentrate on the unification of the long - term x - ray luminosity and the rotational evolution of these neutron star systems in a single picture . with the assumption that these sources are evolving in vacuum with dipole torques ,
their surface dipole fields , b , inferred from the dipole torque formula range from @xmath3 g for ccos to more than @xmath4 g for axp / sgrs . in this picture , the diversity of evolutionary paths is attributed to the differences in the initial dipole and the crustal toroidal fields of the sources ( kaspi 2010 ) .
sources with magnetar dipole fields ( @xmath5 g ) are posited to go through a rapid field decay , which is required for the model to explain the recently discovered so - called low - b magnetars(turolla et al .
2011 , rea et al . 2012 ) .
evolution of the x - ray luminosities of these sources , which are much higher than the rotational powers , are suggested to be governed by the field decay history of the neutron stars depending on the initial crustal and dipole field strengths ( vigano et al .
2013 ) . in this model ,
the apparently missing link between ccos and other classes was suggested to be due to field burial to the crust by the accretion of the supernova matter in the early phase of evolution ( vigano & pons 2012 ) .
the timescale for the subsequent reemergence and the growth of the field to its original strength is estimated to be @xmath6 yr depending on the initial conditions by the same authors .
the dipole field s decay is supposed to proceed after the initial burial and re - emergence .
if there are fallback disks around these systems the picture is rather different .
it was proposed by alpar ( 2001 ) that the source properties , in particular the period clustering , could be explained by the presence ( or absence ) and the properties of fallback disks around these systems .
fallback disk models do not involve burial , re - emergence or decay of the dipole field , which is taken to be constant at its initial value .
neutron stars with fallback disks and conventional dipole magnetic fields @xmath7 g ) could evolve into the x - ray luminosity , period and period derivative range of axp / sgrs on timescales of @xmath8 a few @xmath9 yr ( ertan et al . 2009 ) .
the infrared ( ir ) and optical emission properties of these sources in the quiescent state are consistent with the emission from an irradiated active disk ( ertan et al .
2007 , ertan & 2006 ) .
the model fits to optical and ir data also constrain the dipole field strength to below @xmath10 g on the surface of the star ( ertan et al .
recently discovered low - b magnetars sgr 0418 + 5729 ( rea et al .
2013 ) and swift j1822.3@xmath111606 ( scholz et al . 2012 ) clearly showed that the sgr bursts do not require magnetar strength dipole fields .
this suggests that the sgr bursts could be powered by the quadrupole or higher multipole fields which are localized close to the surface and do not affect the rotation history of the star .
the inner disk applies torque on the star through interaction with the large - scale dipole component of the magnetic field .
the x - ray luminosity and the rotational properties of the low - b magnetarsand the so - called high - b radio pulsar psr j1734@xmath113333 , including its anomalous braking index n = 0.9 @xmath12 0.2 ( espinoza et al .
2011 ) , can also be reproduced by the same model and with @xmath13 g. as an independent and complementary support for this model , recent analyses show that the high energy spectra of axp / sgrs could be produced in their accretion columns consistently with the accretion rates of these sources inferred from the x - ray luminosities ( trmper et al . 2010 , 2013 ; kylafis , trmper & ertan 2014 ) .
these results obtained from the application of the same model with similar basic parameters to sources with rather different observational properties encourages us to investigate the evolution and possible evolutionary links between axp / sgrs ( see mereghetti 2008 ; 2011b for recent reviews ) and other young neutron star populations .
this will help us understand the differences in the initial conditions that lead to emergence of different classes of neutron stars .
in particular , understanding whether there is a relation between the disk masses and the dipole field strengths requires detailed investigations of axp / sgrs and other young neutron star populations . in the present work ,
we concentrate on the dim isolated neutron stars ( xdins ) . at present , there are seven known dim isolated neutron stars ( haberl 2007 ; turolla 2009 ) .
all these sources lie within a distance of @xmath14 pc .
since the solar neighborhood is part of the gould belt , a ring of young stellar systems , the relatively high rate of supernova events in the gould belt should be taken into account in estimating the birth rate of xdins . in the vicinity of the sun ,
about two thirds of the neutron stars are born in the gould belt , while the remaining fraction belongs to the galactic disk ( grenier 2000 ) .
assuming the ages of xdins are @xmath15 myr , simple statistical calculations give a galactic birth rate of @xmath16 century@xmath17 ( popov , turolla & possenti 2006 ) .
the thermal x - ray luminosities of xdins are in the @xmath18 erg s@xmath17 range .
the ages corresponding to these luminosities on the theoretical cooling curves are a few @xmath9 yr . kinematic ages of four xdins estimated from the space velocities and likely birth places are , with large uncertainties , in the range of @xmath19 myr ( motch et al .
2009 ; tetzlaff et al .
2010 ; mignani et al .
these kinematic ages and the cooling ages estimated from the x - ray luminosities of xdins , imply that the characteristic ages , estimated assuming isolated pulsar spin - down , @xmath20 myr , could be significantly greater than the true ages of these sources .
xdins do not show pulsed radio emission .
weak radio emission from two sources were reported ( malofeev , malov & teplykh 2007 ) , but not confirmed yet ( turolla 2009 ; haberl 2007 ) .
strength of the surface dipole fields of xdins inferred from the dipole torque formula are @xmath21 g. if these sources are indeed evolving in vacuum with dipole torques , many of them should be active as radio pulsars .
nondetection of radio pulses from these sources might be due to narrow beaming angles of long - period system .
nevertheless it is not clear why a large population of galactic xdins with rotational properties of the known sources do not show up as radio pulsars .
alternatively , these sources may not have sufficiently strong dipole fields for radio emission .
our results support the latter possibility . in the present work ,
we try to explain the long - term evolution of xdins in the frame of the fallback disk model as applied earlier to axp / sgrs and psr j1734@xmath113333 .
we also discuss the radio properties of xdins based on the predictions of our evolutionary model .
we briefly describe the model in section 2 .
the results of the simulations are given in section 3 .
we summarize our conclusions in section 4 .
we use the code developed to investigate the long - term evolution of axps and sgrs ( see ertan & erkut 2008 ; ertan et al . 2009 ; alpar et al .
2011 ; et al .
2013 for details and applications ) .
we examine the period , the period derivative and the total x - ray luminosity evolution of the model sources , tracing the initial conditions , namely the initial period , @xmath22 , strength of the magnetic dipole field on the pole of the star , @xmath1 , and the initial disk mass , @xmath23 .
in addition to these initial parameters , the evolution is also affected by the irradiation efficiency @xmath24 and the minimum critical temperature , @xmath25 , for the disk to be viscously active . the magneto - rotational instability
( balbus & hawley 1991 ) which generates the turbulent viscosity needed for the disk to transport mass and angular momentum will not work at temperatures below @xmath25 , because the ionization fraction becomes too small .
the radius of the disk could be written as ( g m)^-1/7 ^4/7 ^-2/7 ( lamb , pethick & pines 1973 , davidson & ostriker 1973 ) where @xmath26 is the gravitational constant , @xmath27 and @xmath28 are the mass and magnetic dipole moment of the neutron star , and @xmath29 is the mass - flow rate arriving at the inner disk radius , @xmath30 . when @xmath31 is less than the light cylinder radius , @xmath32 , we take @xmath33 .
accretion will take place in this regime .
typically for the sources we consider , in the accretion phase @xmath34 and the inner disk radius is greater than the co - rotation radius , @xmath35 .
the star is in the propeller phase , spinning down under the disk torques while accreting . over the long - term evolution
, @xmath36 increases with decreasing angular frequency of the neutron star , @xmath37 , while @xmath31 increases with decreasing @xmath29 .
when @xmath31 calculated by equation ( 1 ) is found to be greater than the current value of @xmath36 , we set @xmath38 . in this tracking phase ,
we assume that there is no accretion onto the neutron star .
we solve the diffusion equation for an extended disk with an initial surface - density distribution in the power - law form , @xmath39 , which is the characteristic surface - density profile for steady thin accretion disks ( see e.g. frank , king & raine 2002 ) . at a given time , the temperature in the disk decreases with increasing radial distance @xmath40 from the center . at a given @xmath40 ,
the temperature decreases with time because of fallback disk evolution and decreasing irradiation strength.in the calculations , for numerical reasons , we start with @xmath41 cm .
after the first time step , the radius at which the disk temperature is as low as the minimum temperature @xmath25 for the disk to be active is identified as the dynamical outer radius , @xmath42 , of the viscously active disk .
as the temperatures , starting from the outermost disk , decrease below @xmath25 , @xmath42 propagates inward .
the x - ray irradiation of the disk is important in the evolution of the neutron star , since it prolongs the lifetime of the disk by delaying the inward propagation of @xmath42 .
x - ray irradiation flux can be written as @xmath43 ) ( shakura & sunyaev 1973 ) where @xmath44 is the speed of light . from the results of our earlier work on the ir and x - ray emission of persistent axp / sgrs ( ertan & 2006 )
, we found the value of the irradiation efficiency @xmath24 to be in the range of @xmath45 which is similar to the range of @xmath24 estimated for low - mass x - ray binaries ( dubus et al .
there is a degeneracy between the parameters @xmath24 and @xmath25 . with two extreme values of @xmath24 ,
similar evolutionary curves can be obtained by changing @xmath25 only by a factor of @xmath46 .
the range of @xmath24 obtained in our earlier work constrains @xmath25 to below @xmath47 k ( ertan et al . 2009 ) , consistent with the results of inutsuka & sano ( 2005 ) .
accretion onto the surface of the neutron star is the dominant source of the x - ray luminosity .
the accretion rate , @xmath48 , is related to the x - ray luminosity through @xmath49 where @xmath50 is the radius of the neutron star . in the fallback disk model of axp / sgrs ,
the sources are in the propeller regime and a fraction of the matter arriving at the inner disk radius is accreted onto the star ( alpar 2001 ; chatterjee , hernquist & narayan 2000 ) .
this can be written as @xmath51 where @xmath52 .
for simplicity , we take @xmath53 in all our calculations .
similar results can be obtained with @xmath54 .
depending on the initial conditions , the sources can enter the accretion phase at different epochs of the evolution .
some sources may never enter the accretion phase and probably continue their evolution as radio pulsars until their rotational power is no longer sufficient to produce pulsed radio emission .
other sources , with different initial conditions , may evolve through an initial pulsar phase into a subsequent accretion epoch .
when accretion is not possible , that is , when the inner disk can not penetrate the light cylinder , the x - ray emission is mainly due to intrinsic cooling of the star . for the cooling luminosity , @xmath55
, we use the theoretical cooling curves calculated for the neutron stars with conventional magnetic dipole fields ( page 2009 ) . in the luminosity calculation , in addition to @xmath55 we also include the intrinsic dissipative heating of the neutron star under the dipole and disk torques acting on the star ( alpar 2007 ) .
we use the @xmath56 prescription for the kinematic viscosity @xmath57 where @xmath58 and @xmath59 are the local sound speed and the pressure scale - height of the disk respectively ( shakura & sunyaev 1973 ) . in the long - term evolution of the disk , the mass - flow rate from the outer to the inner disk
is determined by the viscosities in the cold outer disk . following the results of the detailed work on the x - ray enhancement light curves of axp / sgrs ( & ertan 2012 ) we take @xmath60 .
the disk could remain stable ( i.e. , not blown away by radiation pressure ) when @xmath61 for a certain range of @xmath30 values depending on the angle @xmath62 between the magnetic dipole and the rotation axes of the neutron star ( eki & alpar 2005 ) .
the maximum @xmath30 for the disk to remain stable is a few @xmath36 for large @xmath62 , while the disk is stable for all @xmath30 values if the axes are aligned ( eki & alpar 2005 ) .
even if the disk remains stable , it can not apply an efficient torque on the star if the inner disk loses contact with the closed field lines .
an alternative and stronger possibility in this regime is that the inner disk matter can not be propelled efficiently from the system and piles up at the inner disk . in this case , because of the increasing surface - density gradient at the inner disk , @xmath30 approaches to @xmath36 and a steady state is reached with an inner disk that remains in contact with the closed dipole field lines .
the system then follows a tracking phase with @xmath63 ( ertan et al .
2009 ; alpar et al .
2011 ; et al . 2013 ) . as in our earlier work
, we adopt that the inner disk follows this tracking phase when the inner disk can not penetrate the light cylinder . in the accretion regime ,
the disk spin - down torque acting on the star can be written as @xmath64 where the dimensionless torque @xmath65 is a function of the fastness parameter @xmath66 where @xmath67 is the angular frequency of the disk at @xmath68 . from our earlier analysis , we found that the dimensionless torque in the form @xmath69 can produce the period evolution of axp / sgrs with @xmath70 and @xmath71 ( ertan & erkut 2008 ; ertan et al .
2009 ) . substituting @xmath65 in the torque equation
, we have n = ( g m ) ^1/2 ( 1-^2 ) = i where @xmath72 is the moment of inertia of the neutron star .
when @xmath73 , we calculate @xmath74 with @xmath33 using equation ( 1 ) .
when @xmath31 is found to be greater than @xmath36 , we take @xmath38 . in
the spin - down with accretion phase ( @xmath75 ) , substituting equation ( 1 ) in equation ( 2 ) , the period derivative , @xmath76 , of the neutron star is found to be independent of both @xmath48 and @xmath77 when @xmath78 is sufficiently beyond unity .
this is probably the case for most of the persistent axps and sgrs currently in the accretion phase except for a few sources which are likely to be close to rotational equilibrium .
the low - b magnetars sgr 0418 + 5729 and swift j1822.3@xmath111606 , which seem to have completed the accretion epoch , are evolving in the tracking phase with @xmath79 ( alpar et al .
2011 ; benli et al .
2013 ) . in this phase
, the disk torque is proportional to @xmath29 .
the ratio of the ram pressure to magnetic pressure at @xmath36 , and the disk torque decrease with time . by the time the magnetic dipole torque dominates the disk torque ,
the x - ray cooling luminosity of the source is already below the detection limits in most cases . in the tracking phase following the accretion phase
, the pulsed radio emission rate depends on the dipole field strength and the period of the source ( see sections 3 and 4 ) .
we assume that magnetic field decay is negligible within the observable timescale of the axp / sgrs and xdins ( @xmath80 yr ) .
sources with appropriate initial conditions could start their evolution in the tracking phase .
these are likely to be observed as radio pulsars until they cross the radio pulsar death line .
out of seven xdins , six sources have measured period and period derivatives s ) and a tentative ( 2 @xmath81 ) period derivative @xmath82 s s@xmath17 ) was reported for the seventh source rx j1605.3 + 3249 ( pires et al . 2014 ) .
we intend to investigate the evolutionary possibilities of this source when the period derivative is confirmed . ] . for comparison with our model results ,
we have converted the reported x - ray fluxes of the six xdins with known @xmath77 and @xmath76 into the unabsorbed fluxes and luminosities using the distances , @xmath83 values and blackbody temperatures given in the corresponding papers .
there are large uncertainties in the distances of some xdins which are reflected in the uncertainties of the x - ray luminosities .
corrections to luminosities could require modification of our model parameters reported here . for rx j0420.0 - 5022 ,
@xmath84 s ( haberl et al . 2004 ) and @xmath85 s s@xmath17 ( kaplan & van kerkwijk , 2011 ) . the distance of this source was estimated as @xmath86 pc by posselt et al .
the @xmath87 kev observed flux was reported as @xmath88 erg s@xmath17 @xmath89 ( haberl et al .
2004 ) . using the properties of their spectral fit , we calculate the bolometric luminosity as @xmath90 erg s@xmath17 .
the period of rx j1308.6 + 2127 ( rbs 1223 ) is @xmath91 s ( haberl 2004 ) .
the most recent timing analysis gives @xmath92 s s@xmath17 ( kaplan & van kerkwijk 2005 ) . the distance is estimated to be @xmath93 pc by schwope ( 2005 ) and 380 @xmath94 pc by hambaryan et al .
( 2011 ) . using @xmath95 pc and the observed flux reported by schwope , schwarz & greiner ( 1999 )
, we obtained @xmath96 erg s@xmath17 . for rx j0806.4 - 4123 , @xmath97 s and @xmath98 s s@xmath17 ( kaplan & van kerkwijk 2009a ) . the distance of this source was estimated as 235 - 250 pc by posselt et al .
( 2007 ) . using @xmath99 pc and with the observed x - ray flux reported by haberl et al .
( 2004 ) , we find @xmath100 erg s@xmath17 . .
the magnitudes of the magnetic dipole field on the pole of the star , @xmath1 , are given in units of @xmath101 g in the top panel ( see the text for a discussion ) . ]
the period of rx j1856.5 - 3754 is 7.055 s ( tiengo & mereghetti 2007 ) and @xmath102 s s@xmath17 ( van kerkwijk & kaplan 2008 ) .
the distance is highly uncertain .
estimates range from 123@xmath103 pc ( walter et al .
2010 ) to 167@xmath104 pc ( kaplan , van kerkwijk & anderson 2007 ) .
posselt et al .
( 2007 ) estimate the distance as 135@xmath10525 pc . with @xmath106 pc
, @xmath107 is found between @xmath108 erg s@xmath17 and @xmath109 erg s@xmath17 ( walter & lattimer 2002 ; pons et al .
2002 ; drake et al . 2002 ; burwitz et al .
we take @xmath107 = @xmath110 erg s@xmath17 obtained by burwitz et al .
( 2003 ) with the lowest @xmath111 . for rx j2143.0 + 0654 ( rbs 1774 )
, the timing analysis by kaplan & van kerkwijk ( 2009b ) gives @xmath77 = 9.428 s and @xmath112 s s@xmath17 .
the upper limit to the distance was given as 300 pc by posselt , neuhuser & haberl ( 2009 ) and as 390@xmath11430 pc by posselt et al .
the unabsorbed x - ray fluxes reported by kaplan & van kerkwijk ( 2009b ) , zane et al .
( 2005 ) and rea et al .
( 2007 ) are 4.8 , 6.1 and 5.6 @xmath113 erg s@xmath17 @xmath89 respectively . using the flux reported by rea et al .
( 2007 ) , we find @xmath114 erg s@xmath17 with d = 400 pc .
the period of rx j0720.4 - 3125 is @xmath77 = 8.39 s ( haberl et al . 1997 ) .
the most recent analyses give @xmath115 s s@xmath17 ( van kerkwijk et al .
2007 ; hohle et al . 2010 ) .
the distance was estimated as 235 @xmath11 270 pc by posselt et al .
( 2007 ) and as 360@xmath116 by kaplan , kerkwijk & anderson ( 2007 ) .
we use d = 270 pc in our calculations , since it agrees with both estimates .
the x - ray flux was reported as ( 9 @xmath117 erg s@xmath17 @xmath89 by kaplan et al .
( 2003 ) , which corresponds to a bolometric luminosity of @xmath118 erg s@xmath17 .
in the present work , we have investigated the long - term evolution of these six xdins ( table 1 ) . the present day x - ray luminosities , periods and period derivatives of these sources
can be produced simultaneously by the neutron star s evolution with a fallback disk . for a neutron star with a magnetic dipole field @xmath119 g to evolve into the observed period range of xdins ( @xmath120 s ) the system should pass through an accretion phase .
our results indicate that these sources slowed down to long periods during accretion epochs lasting from a few @xmath121 yr to a few @xmath9 yr .
they completed their accretion phases in the past , and now they are in the tracking phase , still spinning down by disk torques , which are weaker than in the accretion epoch . at present , the x - ray luminosities of these sources are likely to be powered by the intrinsic cooling of the neutron star . if the inner disk had never penetrated into the light cylinder the disk torque could not have spun down a neutron star with a conventional @xmath119 g dipole field to xdin periods within the cooling timescale , @xmath122 yr ( e.g. page 2009 ) .
= 1.2 @xmath123 10@xmath124 g ) for all sources .
the disk masses are given in units of @xmath125 in the top panel . ]
we present illustrative model curves in figures 1 and 2 to show the dependence of evolutionary paths on the disk mass , @xmath23 , and the dipole field strength on the pole of the star , @xmath1 , respectively .
different phases of evolution can be followed from the @xmath76 curves given in the bottom panels of figures 1 and 2 . in the accretion phase ( @xmath126 ) , from equation ( 1 ) , it is found that @xmath76 is proportional to @xmath127 , and independent of @xmath29 and @xmath77 .
that is , the evolutionary phases with constant @xmath76 correspond to accretion epochs . in this phase , for a given @xmath23 , the sources with higher @xmath1 reach longer periods .
after the accretion phase , the system enters the tracking phase in which @xmath79 .
the disk torque decreases with decreasing @xmath29 ( the mass inflow rate arriving at @xmath79 ) .
accretion onto the neutron star has stopped , @xmath128 , and the x - ray luminosity is supplied by the intrinsic cooling of the neutron star . in figure 1 , we also give illustrative model curves that could represent the evolution of axp / sgrs ( the two upper curves ) . these model sources with relatively high @xmath1 values
can not acquire the properties of xdins .
our results indicate that xdins could be distinguished from axp / sgrs by having weaker @xmath1 fields . in figure 2 , we give the evolutionary curves for different disk masses keeping the initial dipole magnetic field constant at @xmath129 g. it is seen that systems with greater @xmath23 have longer accretion phases .
we obtain these illustrative model curves with @xmath130 k and @xmath131 . using @xmath132 , evolutionary curves similar to those in figure 2
can be obtained with @xmath133 k. even with the highest reasonable value of @xmath24 , the model curves obtained with @xmath134 k do not lead to xdin or axp / sgr properties for any disk mass . [ cols="^,^,^,^,^,^",options="header " , ] + comparing figures 1 and 2 , it is seen that the dipole magnetic field strength @xmath1 , rather than the disk mass , is the more effective initial condition for determining the long - term rotational evolution of the sources . in figure 1
, @xmath23 is the same for all model sources , and the @xmath1 values vary only by a factor of 4 .
we see that the model curves trace rotational properties of axp / sgr and xdins from the shortest to the longest observed @xmath77 and @xmath76 values . for
a given @xmath1 , different @xmath23 values do not yield significantly different evolutionary paths . in figure 2 ,
disk masses change by an order of magnitude , while the final periods at @xmath135 myr remain in the 2 @xmath11 5 s range .
the @xmath76 curves of the sources with the same @xmath1 but different @xmath23 are also very similar . in the accretion phase ,
the accretion power is the dominant source of the x - ray luminosity , and depends on the initial disk mass @xmath23 . with more detailed analysis
, we also tried to reproduce the individual properties of xdins with known @xmath77 and @xmath76 values .
the evolutionary tracks of these sources are seen in figure 3 .
since all these xdins are currently powered by the intrinsic cooling luminosity of their neutron stars , observed luminosities and the theoretical cooling curves constrain the ages of the model sources .
in all these calculations , we first take @xmath136 and @xmath25 = 100 k. in some cases , we obtained better fits with slightly different @xmath24 and @xmath25 values .
the model curves given in figure 3 are obtained with @xmath25 values that remain in a narrow range between 80 and 110 k. assuming that axp / sgrs and xdins have similar disk compositions , we expect that they have similar critical temperatures
. the irradiation efficiency might change with the accretion rate ; it is expected to be similar for sources in the same accretion regime .
we perform simulations tracing all possible values of @xmath23 and @xmath1 with @xmath137 ms . for the model sources that can enter the accretion phase , the source properties at the end of this phase are not sensitive to @xmath22 ( ertan et al .
for all xdins , the @xmath1 values that can produce the reasonable evolutionary curves remain in the @xmath138 g range .
the ranges of model parameters that can produce the individual properties of the six xdins are listed in table 1 . and @xmath1 that can produce the source properties ( see table 1 ) . here , we show illustrative model curves that can represent the long - term evolution of these xdins .
the values of @xmath1 and @xmath23 used in the models are given in the top panel in units of @xmath101 g and @xmath125 respectively . ]
values inferred from the dipole torque formula .
vertical bars show the ranges of @xmath1 that can produce the properties of the sources in the fallback disk model ( see figure 3 ) .
it is seen that all these sources remain below the lower boundary of the death valley ( death line ) indicating that these xdins can not be normal radio pulsars , if they are evolving with fallback disks . ]
our results imply that all six xdins completed the long - term accretion phase .
they are not accreting matter from the disk at present , while they are still being slowed down by the disk torques . without accretion , these sources are free to emit pulsed radio emission .
however , the dipole fields @xmath1 inferred in the fallback disk model are significantly weaker than those inferred from the dipole torque formula , so that at present the xdins do not have sufficient voltages to sustain pulsed radio emission .
allowed ranges of @xmath1 values for each xdin are given in table 1 and plotted in figure 4 .
all these sources are indeed well below the lower border of the radio pulsar death valley in the extended galactic population , and are not expected to show radio pulses .
younger xdins with shorter periods should be able to emit radio pulses , but they would still be evolving in the long - term accretion phase during which mass - flow onto the neutron star hinders the radio emission . once the accretion epoch is over ,
xdins have weak rotational power and are not likely to produce radio emission .
the inner disk conditions of xdins in the efficient propeller phase could be rather different from those of persistent axps in the accretion phase , for which the boundary layer extends down to the co - rotation radius in a steady state with magnetic heating in the boundary layer negligible compared to viscous dissipation ( rappaport , fregeau & spruit 2004 ) .
by contrast , in the propeller phase , magnetic heating in the boundary layer could be a significant extra heating source in addition to viscous dissipation and irradiation .
this additional heating in the propeller phase can supply the difference ( @xmath139 orders of magnitude ) between the observed r band flux ( kaplan et al .
2011 ) and the model calculation , for an irradiated face on thin disk .
a calculation of the magnetic heating is beyond the scope of the present work ; a detailed study is in progress and will be presented in a separate paper .
we have examined the long - term evolution of the six xdin sources , with measured @xmath76 out of the current total 7 xdins .
first , we find that the fallback disk model is successful in producing the observed individual properties of all these sources .
xdin properties are obtained after the source has gone through a past accretion phase with spin - down , with the fallback disk protruding into the neutron star s magnetosphere . at present , the fallback disk is tracking the light cylinder , with continuing spin - down under the disk torques , but no accretion onto the neutron star .
the x - ray luminosities of xdins are provided by cooling .
next , tracing the initial parameters , we have determined the allowed ranges of the dipole field strength on the pole of the star , @xmath1 , and the disk mass , @xmath23 , for each source ( table 1 ) .
our results constrain the @xmath1 values of the six xdins into the @xmath140 g range . comparing this range of dipole field strength with earlier results obtained by ertan et al .
( 2009 ) for axp / sgrs , we surmise that xdins tend to have dipole fields weaker than those of most known axp / sgrs .
this result should be confirmed through further detailed work on the persistent and transient axp / sgrs .
reasonable evolutionary model curves for xdins can be obtained with a wide range of initial fallback disk masses , @xmath23 , for all sources ( table 1 ) .
we could not test whether there is a correlation between @xmath23 and @xmath1 , since our results do not constrain the disk masses .
for the sources that do not accrete at present , like xdins , it is not possible to constrain the initial mass of the disk in most cases .
a possible @xmath141 correlation could be tested through further investigation of persistent axp / sgrs which are powered by accretion onto the star .
this analysis will help us understand the differences in the initial conditions of different young neutron star populations .
all known xdins are close - by objects within @xmath14 pc of the sun . statistical analysis considering the properties of the gould belt gives a galactic birth rate of @xmath142 1 century@xmath17 ( popov et al .
2006 ; see also section 1 ) .
this raises a critical question for the models : only the xdins in the solar neighborhood are observable in x - rays
. the inferred large galactic population should be detectable in the radio band if they are active radio pulsars .
if these sources evolve in vacuum without fallback disks , their dipole field on the pole of the star can be estimated as @xmath143 g. observed @xmath77 and @xmath76 values give @xmath144 g. it is expected to see these sources as radio pulsars , if xdins indeed slow down by magnetic dipole torques with these strong fields .
five out of seven xdin sources , located within 0.4 kpc , have periods greater than 7 s. among the currently known radio pulsars in the atnf pulsar catalog ( manchester et al .
2005 ) , there is only a single radio pulsar with @xmath145s ( @xmath146 s ) and @xmath147 g which is located at a distance of 3.3 kpc .
we are not sure whether this can be accounted for with the selection effects and even the most conservative beaming fraction formulas for the long - period systems , if xdins and radio pulsars with large dipole magnetic fields are taken to be members of the same population . a detailed population synthesis or statistical analysis
will not be attempted here . this problem encountered in the dipole torque model
is naturally resolved in the fallback disk model . during the initial long - term accretion phase radio emission
is suppressed due to mass - flow onto the star .
is it possible to observe pulsed radio emission from these sources after the mass accretion terminates ?
the neutron stars that evolve like the six xdins we studied here are not likely to produce beamed radio emission when accretion stops , because : ( 1 ) their dipole fields are much weaker than inferred from the dipole torque formula ( table 1 ) , and ( 2 ) at the end of the accretion phase , they have already attained long periods . in figure 3 , accretion phases correspond to the epochs with constant @xmath76 given in the lower panel .
the period at the end of the accretion phase corresponds to a location in the @xmath148 plane which is already below the lower border of the death valley .
subsequent evolution moves the source even further away from the death valley , to the present locations shown in figure 4 . to sum up : ( 1 ) the rotational properties and the x - ray luminosities of xdins
can be explained by the fallback disk model that was employed earlier to explain the general properties of axp / sgrs ( figure 1 ) .
the main disk parameters used in the present work for different xdin sources are similar to each other and also to those used in the long - term evolution models of axp / sgrs ( table 1 ) .
( 2 ) the distinction between xdins and axp / sgrs seems to be that the former have weaker dipole magnetic fields . finally ,
( 3 ) the model can also account for non - detection of many radio pulsars evolving to the properties of known xdins .
a neutron star with a conventional dipole field must have passed through the long - term accretion phase to acquire the periods of known xdins within the cooling timescale . in the accretion phase ,
the radio emission is quenched by the mass flow onto the neutron star .
after termination of the accretion phase , the sources no longer have sufficient rotational power for radio emission .
we acknowledge research support from sabanci university , and from tbitak ( the scientific and technological research council of turkey ) through grant 113f166 . m.a.a . is a member of the science academy - bilim akademisi , istanbul , turkey .
99 alpar m. a. 2001 , apj , 554 , 1245 alpar m. a. 2007 , ap&ss , 308 , 133 alpar m. a. , ertan . , . , 2011 , apj , 732 , l4 balbus s. a. , hawley , j. f. 1991 , apj , 376 , 214 benli o. , . , ertan . , alpar m. a. , trmper j. e. , kylafis n. d. 2013 , apj , 778 , 119 burwitz v. , haberl f. , neuhuser r. , predehl p. , trmper j. , zavlin v. e. , 2003 , a&a , 399 , 1109 chatterjee p. , hernquist l. , narayan r. , 2000 , apj , 534 , 373 chen k. , ruderman m. 1993 , apj , 402 , 264 alikan . , ertan . , alpar m. a. , trmper j. e. , kylafis n. d. , 2013 , mnras , 431 , 1136 alikan . , ertan . , 2012 , apj , 758 , 98 davidson k. , ostiker j. p. , 1973
, apj , 179 , 585 drake j. j. et al .
2002 , apj , 572 , 996 dubus , g. , lasota j. p. , hameury j. m. , charles p. 1999 , mnras , 303 , 139 eki k. y. , alpar m. a. , 2005 , apj , 620 , 390 ertan , alikan .
2006 , apj , 649 , l87 ertan . , erkut m. h. 2008 , apj , 673 , 1062 ertan . , erkut m. h. , eki k. y. , alpar m. a. 2007 , apj , 657 , 441 ertan . , eki k. y. , erkut m. h. , alpar m. a. 2009 , apj , 702 , 1309 espinoza c. m. , lyne a. g. , kramer m. , manchester r. n. , kaspi v. m. , 2011 , apj , 741 , l13 frank j. , king a. , raine d. j. , 2002 , accretion power in astrophysics , cambridge university press , pp .
398 grenier i. a. , 2000 , a&a , 364 , l93 haberl f. , 2004 , advances in space research , 33 , 638 haberl f. , 2007 , ap&ss , 308 , 181 haberl f. , motch c. , buckley d. a. , zickgraf f .- j . , pietsch w. , 1997 , a&a , 326 , 662 haberl f. et al . , 2004 , a&a , 424 , 635 hambaryan v. , suleimanov v. , schwope a. d. , neuhuser r. , werner k. , potekhin a. y. , 2011 , a&a , 534 , 74 hohle m. m. , haberl f. , vink j. , turolla r. , zane s. , de vries c. p. , mndez m. , 2010 , a&a , 521 , 11 inutsuka s. , sano t. , 2005 , apj , 628 , l155 kaplan d. l. , van kerkwijk m. h. 2005 , apj , 635 , l65 kaplan d. l. , van kerkwijk m. h. 2009a , apj , 692 , l62 kaplan d. l. , van kerkwijk m. h. 2009b , apj , 705 , 798 kaplan d. l. , van kerkwijk m. h. 2011 , apj , 740 , l30 kaplan d.l . , van kerkwijk m. h. , marshall h. l. , jacoby b. a. , kulkarni s. r. , frail d. a. , 2003 , apj , 590 , 1008 kaplan d. l. , van kerkwijk m. h. , anderson j. 2007 , apj , 660 , 1428 kaplan d. l. , kamble a. , van kerkwijk m. h. , ho w. c. g. 2011 , apj , 736 , 117 kaspi v. m. 2010 , proceedings of the national academy of science , 107 , 7147 keane e. f. , kramer m. 2008 , mnras , 391 , 2009 kylafis n. d. , trmper j. e. , ertan . , 2014 , a&a , 562 , a62 lamb f. k. , pethick c. i. , pines d. 1973 , apj , 184 , 271 malofeev m. v. , malov o. i. , teplykh d. a. , 2007 , in isolated neutron stars : from the interior to the surface , ed .
s. zane , r. turolla , d. page , springer , berlin heidelberg , new york , manchester r. n. , hobbs g. b. , teoh a. , hobbs m. 2005 , vizier online data catalog , 7245 , 0 mereghetti s. , 2008 , a&arv , 15 , 225 mereghetti s. , 2011a , high - energy emission from pulsars and their systems , astrophysics and space science proceedings , springer - verlag berlin heidelberg , p. 345 mereghetti s. , 2011b , advances in space research , 47 , 1317 mignani r. p.
, 2013 , mnras , 429 , 3517 motch c. , pires a. m. , haberl f. , schwope a. , zavlin v. e. , 2009 , a&a , 497 , 423 page , d. , 2009 , in `` neutron stars and pulsars '' , w. becker , ed . , astrophysics and space science library 357 , 247 pires a. m. , haberl f. , zavlin v. e. , motch c. , zane s. , hohle m. m. , 2014 , a&a , 563 , a50 pons j. a. , walter f. m. , lattimer m. , prakash m. , neuhuser r. , an p. , 2002
, apj , 564 , 981 popov s. b. , turolla r. , possenti a. , 2006 , mnras , 369 , l23 posselt b. , neuhuser r. , haberl f. , 2009 , a&a , 496 , 533 posselt b. , popov s. b. , haberl f. , trmper j. , turolla r. , neuhuser r. , 2007 , ap&ss , 308 , 171 rappaport s. a. , fregeau j. m. , spruit h. 2004 , apj , 606 , 436 rea n. et al . , 2007 , mnras , 379 , 1484 rea n. et al . 2012 , apj , 754 , 27 rea n. et al .
, 2013 , apj , 775 , l34 scholz p. , ng c .- y . , livingstone m. a. , kaspi v. m. , cumming a. , archibald r. f , 2012 , apj , 761 , 66 schwope a. d. , schwarz r. , greiner j. , 1999 , a&a , 348 , 861 schwope a. d. , hambaryan v. , haberl f. , motch c. , 2005 , a&a , 441 , 597 shakura n. i. , sunyaev r. a. , 1973 , a&a , 24 , 337 tetzlaff n. , neuhuser r. , hohle m. m. , maciejewski g. , 2010 , mnras , 402 , 2369 tiengo a. , mereghetti s. 2007 , apj , 657 , l101 trmper j. e. , zezas a , ertan . , kylafis n. d. , 2010 , a&a , 518 , a46 trmper j. e. , dennerl k. , kylafis n. d. , ertan . , zezas a. 2013 , apj , 764 , 49 turolla r. , 2009 , in `` neutron stars and pulsars '' , w. becker , ed . , astrophysics and space science library 357 , 141 turolla r. , zane s. , pons j. a. , esposito p. , rea n. 2011 , apj , 740 , 105 van kerkwijk m. h. , kaplan d. l. , 2008 , apj , 673 , l163 van kerkwijk m. h. , kaplan d. l. , pavlov g. g. , mori k. 2007 , apj , 659 , l149 vigan d. , pons j. a. 2012 , mnras , 425 , 2487 vigan d. , rea n. , pons j. a. , perna r. , aguilera d. n. , miralles j. a. 2013 , mnras , 434 , 123 walter f. m. , lattimer j. m. 2002 , apj , 576 , l145 walter f. m. , eisenbeis t. , lattimer j. m. , kim b. , hambaryan v. , neuhuser r. , 2010 , apj , 724 , 669 zane s. , cropper m. , turolla r. , zampieri l. , chieregato m. , drake j. j. , treves a. , 2005 , apj , 627 , 397 | the dim isolated neutron stars ( xdins ) have periods in the same range as the anomalous x - ray pulsars ( axps ) and the soft gamma - ray repeaters ( sgrs ) .
we apply the fallback disk model , which explains the period clustering and other properties of axp / sgrs , to the six xdins with measured periods and period derivatives .
present properties of xdins are obtained in evolutionary scenarios with surface dipole magnetic fields @xmath0 g. the xdins have gone through an accretion epoch with rapid spin - down earlier , and have emerged in their current state , with the x - ray luminosity provided by neutron star cooling and no longer by accretion .
our results indicate that the known xdins are not likely to be active radio pulsars , as the low @xmath1 , together with their long periods place these sources clearly below the death valley .
0minp_0 , 18sgr 0418 + 5729 42axp 0142 + 61 [ firstpage ] pulsars : individual ( axps ) stars : neutron x - rays : bursts accretion , accretion disks |
let @xmath0 be a qbf in _ prenex cnf ( pcnf ) _ where @xmath1 with @xmath2 is the prefix containing quantified propositional variables @xmath3 and @xmath4 is a quantifier - free cnf . given a pcnf @xmath0 , an _
unsatisfiable core ( uc ) _ of @xmath5 is an unsatisfiable pcnf @xmath6 such that @xmath7 and @xmath8 .
the prefix @xmath9 is obtained from @xmath10 by deleting the quantified variables which do not occur in @xmath11 .
a _ _ minimal unsatisfiable core ( muc ) _ _ of @xmath5 is an unsatisfiable core @xmath12 of @xmath5 where , for every @xmath13 , the pcnf @xmath14 is satisfiable . _
incremental solving _ is crucial for the computation of mucs in the context of propositional logic ( sat ) , e.g. @xcite .
modifications of a cnf by adding and deleting clauses in incremental solving are typically implemented by _ selector variables _ and _ assumptions _ @xcite .
an added clause @xmath15 is augmented by a fresh selector variable @xmath16 so that actually @xmath17 is added . via the solver api
, the user assigns these variables as assumptions under which the cnf is solved to control whether a clause is effectively present in the cnf .
different from the assumption - based approach , the sat solver zchaff @xcite provides an api to modify the cnf by adding and removing _ groups _ ( sets ) of clauses .
clauses are associated with an integer i d of the group they belong to . in assumption - based incremental
solving , clause groups may be emulated by augmenting all clauses in a group by the same selector variable .
the user must specify the necessary assumptions via the api in all forthcoming solver invocations to enable and disable the right groups .
in contrast to that , zchaff allows to delete groups by a single api function call . in terms of usability , we argue that incremental solving by a clause group api is less error - prone , more accessible to inexperienced users , and facilitates the integration of the solver in other tools .
we present a novel clause group api of our qbf solver depqbf ( version 4.0 or later ) in the style of zchaff .
different from zchaff , we implemented clause groups based on selector variables and assumptions to combine the conceptual simplicity of zchaff s api with state of the art assumption - based incremental solving . as a novel feature of our api , the handling of selector variables and assumptions
is entirely carried out by the solver and is hidden from the user .
our approach is applicable to any sat or qbf solver supporting assumptions .
based on the novel clause group api of depqbf , we implemented a tool to compute mucs of pcnfs , a problem which has not been considered so far .
results on benchmarks used in the qbf gallery 2014 illustrate the applicability of the clause group api for muc computation of pcnfs .
depqbf is a solver for pcnfs based on the qbf - specific variant of the dpll algorithm @xcite with learning @xcite . since version 3.0 @xcite
, depqbf supports incremental qbf solving via an api to add and remove clauses in a stack - based way ( cf .
fig . 3 in @xcite ) .
this api is suitable for solving incremental encodings where clauses added most recently tend to be removed again in subsequent solver calls , like reachability problems such as conformant planning @xcite or bounded model checking @xcite .
the new clause group api of depqbf , however , allows to add and delete clauses _ arbitrarily _ , which is necessary for the incremental computation of mucs of pcnfs .
we first present our novel approach to keeping selector variables invisible to the user , which is a unique feature of depqbf . to this end , we distinguish between selector variables and variables in the encoding .
let @xmath18 be a sequence of pcnfs .
we consider variables over which the pcnfs @xmath19 are defined as _ user variables _ because they are part of the problem encoding represented by @xmath20 . when solving @xmath20 incrementally , _
selector variables _ used to augment clauses in @xmath19 are not part of the original encoding .
variables @xmath21 are stored in an array @xmath22 indexed by an integer i d @xmath23 of @xmath21 such that @xmath24 = v$ ] .
user and selector variables reside in separate sections of @xmath22 : | we consider the incremental computation of minimal unsatisfiable cores ( mucs ) of qbfs . to this end , we equipped our incremental qbf solver depqbf with a novel api to allow for incremental solving based on clause groups .
a clause group is a set of clauses which is incrementally added to or removed from a previously solved qbf .
our implementation of the novel api is related to incremental sat solving based on selector variables and assumptions .
however , the api entirely hides selector variables and assumptions from the user , which facilitates the integration of depqbf in other tools .
we present implementation details and , for the first time , report on experiments related to the computation of mucs of qbfs using depqbf s novel clause group api . |
after the discovery of the higgs - like particle with mass around 125 gev by the atlas and cms collaborations at cern , still many convincing reasons exist to believe that the resolution of the hierarchy problem lies in new physics around the tev mass scale .
in fact , there are at least two , not necessarily mutually exclusive scenarios , offered as solutions of the hierarchy problem : low energy supersymmetry at around 1 tev . large extra dimensions and a low scale for quantum gravity at around 1 tev @xcite .
here we discuss some rather generic features about string and brane compactifications , which are relevant for the search of new physics at the lhc and which might provide concrete and fundamental realizations of the general scenarios mentioned above . in fact , the relevance of strings for the experiments at the lhc is closely linked to the question what is the typical scale of string theory , called the string mass @xmath0 .
a priori , considering d - brane compactifications with open strings on branes , the string scale is a free parameter , not being determined by the standard model gauge interactions .
hence in this class of models @xmath0 can be in principle everywhere between the weak scale and the planck scale .
therefore we divide the discussion into two cases , namely into low and high string scale compactifications .
the most spectacular and most direct lhc signatures from strings will emerge in case @xmath0 is very low , namely at the order of tev
. then the gravitational and gauge interactions are unified at around 1 tev , and the observed weakness of gravity at lower energies is due to the existence of large extra dimensions , where the volume @xmath1 of the internal space in units of @xmath0 must be of the order @xmath2 .
this scenario can be nicely realized by d - brane compactifications ( for a review see e.g. [ ] ) . here
gravitons , i.e. closed strings may scatter into the extra space and by this the gravitational coupling constant is decreased to its observed value . on the other hand , the particles of the standard model correspond to the zero modes of open strings that are confined to lower dimensional d - branes , which typically fill part of the extra - dimensions and also intersect among each other in the higher dimensional space .
a typical brane configuration@xcite is shown in fig.1.@xcite gauge bosons emerge from open strings ending on a the same stack of overlapping branes , with u(1 ) gauge bosons ending on the same brane , while non - abelian gauge bosons like gluons stretch between distinct branes in a single stack .
quarks and leptons emerge from strings stretching between intersecting stacks of branes .
moreover a whole tower of massive open string excitations will open up at around @xmath3 , where the new particles follow the well known regge `` trajectories '' of vibrating strings , @xmath4 with spin @xmath5 particles populating lines with the universal slope @xmath6 determined by the fundamental string mass scale @xmath7 , and `` intercepts '' @xmath8 . at the @xmath9-th level
, the spins of excited gluons range from 0 to @xmath10 and of excited quarks from 1/2 to @xmath11 .
the standard model spectrum is replicated at each level @xmath9 by higher spin particles .
let us list what kind of string signatures from a low string scale and from large extra dimensions in brane models can be possibly expected at the lhc : 0.3 cm \(i ) the discovery of string regge excitations with masses of order @xmath0 .
as we will explain in the following , the production cross sections and the decay rates of the heavy string excitations can be computed in a model independent way for a large class of d - brane models , such that universal predictions for the discovery of these particles can be made .
in fact , the production of string regge excitations will lead to new contributions to standard model scattering processes , like qcd jets or scattering of quarks into leptons or gauge bosons , which can be measurable at lhc in case the string scale is low @xcite .
0.3 cm \(ii ) the discovery of new exotic particles around @xmath0 .
another very interesting signal for new stringy physics at the lhc is the production of heavy neutral @xmath12 gauge bosons ( see e.g. [ ] ) .
these particles are quite generic in any string compactification , and they receive their mass via a green - schwarz mixing with axionic scalar fields .
e.g. in a four stack d6-brane model with gauge group @xmath13 , see fig.1 , at least one extra @xmath14 gauge bosons will get a mass of order @xmath15 by the green - schwarz effect .
0.3 cm \(iii ) the discovery of ( non - perturbative ) quantum gravity effects in the form of mini black holes .
one of the most exciting possibilities for the lhc is the discovery of small higher - dimensional black holes that can be formed when two sufficiently energetic particles collide @xcite .
this means that effects of higher dimensional quantum gravity can get strong if the string scale is low around the tev scale , and if the volume of the extra dimensions is large .
the geometrical cross section for the production of mini black holes is of the order @xmath16 where @xmath17 is the black hole mass , i.e. the effective scale of quantum gravity , and @xmath18 for higher dimensional black holes .
since the production of mini black holes is basically a non - perturbative effect , the black hole mass is suppressed by the string coupling constant compared to the string scale : @xmath19 therefore , for weak string coupling , the threshold for black hole production is almost inevitably higher than the threshold for the production of string resonances .
0.3 cm \(iv ) the discovery of kaluza - klein ( kk ) excitations . in superstring theory with d - branes there are two classes of kaluza - klein states : bulk excitations of closed string states like kk graviton etc .
and the excitations associated to the extra dimensions of d - branes extending into the bulk space .
the masses of bulk states are determined by the properties of the six - dimensional calabi - yau manifold .
the kk excitations of d - brane states are charged under the respective gauge group and their masses depend on how d - branes wrap into closed surfaces ( cycles ) inside calabi - yau manifold .
experimental signatures of kk states are common to all extra - dimensional extensions of the standard model and will be discussed elsewhere in this volume .
now let us turn to string compactifications with the string scale much above the tev region . in fact , there are no compelling reasons why the string mass scale should be much lower than the planck mass . in the large volume compactifications of refs .
[ ] it was shown that that one can stabilize moduli in such a way that the string scale @xmath0 is at intermediate energies of about @xmath20 .
then the internal cy volume @xmath1 is of order @xmath21 .
the motivation for this scenario is to obtain a low supersymmetry breaking scale around @xmath22 , since one derives the following relation for the gravitino mass : @xmath23 hence the experimental indications for this kind of scenario at the lhc would be given by usual signatures for low energy supersymmetry , whose details largely depend on further assumptions of the concrete compactification scheme
. finally the string scale can be also around the planck scale or also a bit lower around the gut scale of about @xmath24 gev , a choice , which is motivated by the unification of the gauge coupling constants of the standard model .
however d - brane models with such a high string scale can nevertheless lead to some rather model independent new physics predictions around the tev scale . specifically , as discussed in [ ] , in generic 4-stack d - brane models one typically arrives at a simple extension of the standard model , called sm@xmath25 ( see fig.1 ) obtained by adding to the scalar sector a complex @xmath26 singlet that has the quantum numbers of the right - handed neutrino , @xmath27 , and to the gauge sector an @xmath14 that is broken by the vacuum expectation value of @xmath27 ( see also [ ] ) .
the associated @xmath28-gauge boson can be relatively light .
the lightest string resonances are excited quarks and gluons which should be copiously produced at the lhc energies just above the string threshold . they are produced by gluon - gluon or quark - gluon fusion , at the rates typical for qcd processes .
these resonances decay into gluon pairs or quarks and gluons , giving rise to spectacular resonance bumps in dijet cross sections . as an example , consider the lightest quark resonances @xmath29 , which appear in two separate spin @xmath30 and @xmath31 eigenstates .
they give rise to a peak in the squared modulus of the quark - gluon scattering amplitude , at the center of mass energies close to their mass @xmath0 : @xmath32 , \label{qg}\end{aligned}\ ] ] where @xmath33 and @xmath34 are the mandelstam invariants of this particular subprocess , and @xmath35 is the qcd coupling constant @xmath36 . here
, the resonance poles have been softened to a breit - wigner form with the partial widths @xmath37 , as calculated in ref .
the squared amplitudes for all @xmath38 and @xmath39 parton subprocesses are collected in refs .
[ , ] while the decay rates can be found in ref .
it is remarkable that the amplitude ( [ qg ] ) , as well as the amplitude describing @xmath40 scattering , do _ not _ contain any reference to extra dimensions .
recall that one of the most common criticisms of string theory is that it does not predict the shape or size of extra dimensions responsible for the properties of low energy effective theory usually called the landscape problem but here , this problem has been circumvented by a complete _ universality _ of the amplitudes . as long as quark and gluons
appear in the low energy spectrum , string resonances contribute in a universal way , common to _ all _ compactifications .
there are conceivably no other extensions of the standard model that would produce such unambiguous , spectacular signatures of strongly interacting , higher spin particles . in reality ,
string resonances are very easy to discover .
the amplitudes involving four and more fermions , like @xmath41 , do depend on the details of compactification , however they do not significantly contribute to dijet cross sections for two reasons : 1 ) color triplets yield smaller @xmath42 qcd factors than gluon octets 2 ) at high energies ( large feynman s @xmath43 ) , antiquark lumnosities inside protons are lower than quark and gluon luminosities .
actual dijet mass distributions are obtained by convoluting parton cross sections with the proton structure functions . as seen in fig.2,@xcite @xmath44
is the most favorable channel for discovering string resonances , with an optimal balance of luminosity and color factors .
no significant bumps in dijet mass distributions have been observed yet at the lhc in the initial runs at @xmath45 tev and @xmath46 tev .
the results of cms searches for string resonances , based on the integrated luminosity of 4 fb@xmath47 at @xmath48= 8 tev , are shown on fig.3.@xcite the data are presented as the observed upper limits ( at 95% confidence level ) on @xmath49 where @xmath50 is the branching fraction of @xmath29 into @xmath51 and @xmath52 is the acceptance for the respective kinematic requirements . the cross section @xmath53 is integrated over the mass bins determined by the uncertainty in jet energy resolution .
tev , the width of @xmath29 resonance , @xmath54 gev,@xcite is sufficiently small to justify the narrow resonance approximation applied to cms data . ]
based on these data , cms quote the lower bound , @xmath55 tev , on the string mass .
gev by searching in the top quark plus gluon decay channels.@xcite this provides an additional bound on a class of `` warped '' string compactifications.@xcite ] low mass string theory will get another chance in the next run . as seen in fig.4,@xcite string mass scales as high as 7 or even 8 tev can be reached by lhc operating at its design center of mass energy @xmath56 tev .
no matter what the string scale is , all d - brane models predict the existence of more neutral gauge bosons than the standard model .
the reason is that gauge groups are associated to stacks of d - branes and in this framework , the color @xmath42 group emerges from a stack of at least three branes which , in the minimal case , generate @xmath57 , with the extra @xmath14 coupled to the baryon number @xmath58 . since @xmath58 is an anomalous symmetry , the corresponding @xmath12 gauge boson receives a mass via green - schwarz mechanism , of order @xmath15 . in general , there are more than one extra gauge bosons .
they fall into two classes : anomalous , massive @xmath12 or non - anomalous @xmath59 with the mass generated by vacuum expectation values of an extended higgs sector . in the latter case
, there is no direct relation between the mass and the string scale @xmath0 .
there are many other extensions of the standard model involving extra gauge bosons , so such a particle would constitute at most a `` harbinger '' of string theory .
it is natural to expect that @xmath12 and/or @xmath28 couple to quarks : in this case a bump would be observed in the dijet invariant mass spectrum .
unfortunately , no extra gauge bosons have been found yet at the lhc , see fig.5.@xcite in this case , cms collaboration quotes @xmath60 tev assuming that @xmath12 couples with the same strength as @xmath61 .
if the string mass is too high for the string resonances or @xmath12 to be discovered at the lhc , @xmath28 gauge bosons , with masses determined by an extended higgs sector , still remain as a possibility . in a minimal model of that type , called sm@xmath25,@xcite one additional d - brane intersects the @xmath62 stack .
the non - anomalous @xmath28 couples to the combination of @xmath63 ( @xmath64 is the lepton number ) and the right - handed isospin @xmath65 .
the strength of its @xmath66 coupling depends on the combination , and the existing dijet data can be interpreted in this context to establish more specific bounds on @xmath28 mass , as seen in fig.6.@xcite the lhc , operating at its design center of mass energy , will explore a wider range of @xmath28 masses .
no low mass string or the associated extra gauge bosons have been observed in the lhc runs at @xmath45 tev and @xmath46 tev . unlike in many other beyond the standard model extensions , due to spectacular strength of dijet signals , low mass strings
will be either discovered at the early stage of the next run or more stringent bounds will be established very soon . with the present bound @xmath55 tev ,
the best we can hope for is the discovery of the lowest , @xmath67 string excitations of quarks and gluons because higher levels and the accompanying mini black holes will be most likely beyond the reach of lhc . nevertheless , with such large cross sections
, it should be possible to identify higher spin string excitations of gauge bosons , quarks and leptons .
this work was partially supported by the erc advanced grant strings and gravity " ( grant.no .
32004 ) and by the dfg cluster of excellence
origin and structure of the universe " .
in addition , this material is based in part upon work supported by the national science foundation under grant no .
any opinions , findings , and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the national science foundation .
l. a. anchordoqui , i. antoniadis , h. goldberg , x. huang , d. lst , t. r. taylor and b. vlcek , `` lhc phenomenology and cosmology of string - inspired intersecting d - brane models , '' phys . rev .
d * 86 * , 066004 ( 2012 ) [ arxiv:1206.2537 [ hep - ph ] ] .
l. a. anchordoqui , h. goldberg , d. lst , s. nawata , s. stieberger and t. r. taylor , `` dijet signals for low mass strings at the lhc , '' phys .
* 101 * , 241803 ( 2008 ) [ arxiv:0808.0497 [ hep - ph ] ] .
d. lst , o. schlotterer , s. stieberger and t. r. taylor , `` the lhc string hunter s companion ( ii ) : five - particle amplitudes and universal properties , '' nucl .
b * 828 * , 139 ( 2010 ) [ arxiv:0908.0409 [ hep - th ] ] .
j. p. conlon , c. h. kom , k. suruliz , b. c. allanach and f. quevedo , `` sparticle spectra and lhc signatures for large volume string compactifications , '' jhep * 0708 * ( 2007 ) 061 [ arxiv:0704.3403 [ hep - ph ] ] .
s. iso , n. okada and y. orikasa , `` classically conformal @xmath68 l extended standard model , '' phys .
b * 676 * , 81 ( 2009 ) [ arxiv:0902.4050 [ hep - ph ] ] , `` the minimal b - l model naturally realized at tev scale , '' phys .
d * 80 * , 115007 ( 2009 ) [ arxiv:0909.0128 [ hep - ph ] ] .
b. hassanain , j. march - russell and j. g. rosa , `` on the possibility of light string resonances at the lhc and tevatron from randall - sundrum throats , '' jhep * 0907 * , 077 ( 2009 ) [ arxiv:0904.4108 [ hep - ph ] ] . | we explain what features of string theory can be tested at the lhc .
we review the present bounds on the string mass and on extra gauge bosons as well as the prospects for the upcoming experiments . |
spherical shells of charge carriers appear in a multitude of systems , such as multielectron bubbles in liquid helium @xcite , metal nanoshells coating a non - conducting nanograin @xcite , carbon cages and fullerenes .
although the properties of flat two - dimensional systems have been widely studied , revealing new physics , the properties of _ spherical _ two - dimensional systems are much less well - studied . in this paper
, we investigate the possibility and the properties of cooper pairing in the spherical geometry .
the particular spherical two - dimensional system that we focus on is the multielectron bubble in liquid helium .
when a flat surface of helium is charged with electrons above a critical charge density , an instability occurs with the surface opening to subsume a large number of electrons forming a bubble .
these multielectron bubbles ( mebs ) are typically micron - sized cavities inside liquid helium , containing a nanometer thin film of electrons on the inner surface of the bubble . the cavity is forced open by the coulomb repulsion of the electrons which is balanced by the surface tension of the helium .
the equilibrium shape of the bubble is spherical , with a radius @xmath0 determined by the number of electrons and the pressure on the helium .
the bare single electron states on the surface of the spherical bubble are angular momentum eigenstates and have discrete energies , characterized by the angular momentum @xmath1 and with degeneracy @xmath2 . at low temperature
there will be a well defined fermi surface located at the highest occupied state .
small - amplitude shape oscillations , including surface waves , can be quantized as spherical ripplons .
the electrons can interact with these ripplons , and we will show that this leads to an attractive effective interaction between electrons .
this paper has two distinct but interwoven parts . in the first part ( section ii ) we discuss in detail how the interactions between electrons and ripplons in the bubble can lead to a cooper pairing scenario .
the goal of the first part is to show that a bcs - type hamiltonian provides a plausible description of the electronic system in multielectron bubbles at low temperatures , and to illustrate the relevant values of the parameters of the model hamiltonian . in the second part ( sections iii - iv ) we investigate the general properties of this hamiltonian using the richardson solution for the reduced bcs hamiltonian describing pairing e.g. in nanograins .
both the ground state properties and the density of states of a spherical two - dimensional bcs system , are derived and discussed .
first , we investigate the three ingredients of the full hamiltonian : the electronic part , the ripplonic part , and the electron - ripplon coupling . then , we analyze the effective interaction between electrons , resulting from both the coulomb interactions and the ripplon - mediated electron - electron interaction .
finally , we arrive at a bcs - type hamiltonian for the multielectron bubble .
* the spherical 2d electron system * the well - known hamiltonian of interacting electrons in a flat 2d electron gas ( 2deg ) in the jellium model can be written in second quantization as @xmath3 where @xmath4 create and destroy an electron with wave number @xmath5 and spin @xmath6 , and @xmath7 where @xmath8 is the electron mass , @xmath9 is the surface of the 2d system , @xmath10 is the permittivity of the medium and @xmath11 is the electron charge .
the hamiltonian of the interacting _ spherical _ electronic system has a very similar form , provided that one uses spherical harmonics @xmath12 instead of plane waves as the single - particle basis functions .
for this purpose we use the operators @xmath13 and @xmath14 that create , resp .
, annihilate an electron in the angular momentum eigenstate @xmath15 , i.e. @xmath16 . the hamiltonian of the interacting spherical two - dimensional electron gas ( s2deg ) becomes @xmath17 where we use the notation @xmath18 and @xmath19 where @xmath0 is the radius of the sphere .
the typical scale of the kinetic energy @xmath20 is listed for several bubble sizes and pressures in table i. the operator @xmath21 creates a spin @xmath6 electron in a single particle state resulting from adding the angular momenta @xmath15 and @xmath22 formally , we have@xmath23{c}% \hat{c}_{(l , m)\otimes(j , n),\sigma}^{+}=% % tcimacro{\dsum \limits_{l^{\prime}=|l - j|}^{l+j}}% % beginexpansion { \displaystyle\sum\limits_{l^{\prime}=|l - j|}^{l+j } } % endexpansion% % tcimacro{\dsum \limits_{m^{\prime}=-l^{\prime}}^{l^{\prime}}}% % beginexpansion { \displaystyle\sum\limits_{m^{\prime}=-l^{\prime}}^{l^{\prime } } } % endexpansion \sqrt{\dfrac{(2l+1)(2j+1)}{4\pi(2l^{\prime}+1)}}\left\langle j,0;l,0|l^{\prime } , 0\right\rangle \\ \times\text { } \left\langle j , n;l , m|l^{\prime},m^{\prime}\right\rangle \text { } \hat{c}_{l^{\prime}m^{\prime},\sigma}^{+ } , \end{array}\ ] ] where @xmath24 is the clebsch - gordan coefficient for combining the angular momenta @xmath15 and @xmath25 into a state of angular momentum @xmath26 . the role of the momentum is now taken by the angular momentum .
indeed , taking @xmath27 for large @xmath1 links the result for the spherical case to that for the flat case .
this link was noted previously for the structure factor of the spherical 2d electron gas @xcite , and its response to a weak magnetic field .
when the coulomb energy is small compared to the kinetic energy , the electrons fill up a fermi sea of angular momentum states @xmath1 ( with degeneracy @xmath2 ) up to a fermi level @xmath28 .
the level splitting at the fermi level , @xmath29 , is given in table i for some typical bubbles .
the typical value of the distance between electronic levels is thz ( or mk ) .
note that the @xmath30 term is absent in the coulomb part of the hamiltonian ( [ hel ] ) : this term is exactly cancelled by the surface tension energy of the helium as shown in ref .
@xcite . in hamiltonian ( [ helflat ] ) for the flat 2d electron gas in jellium ,
the @xmath31 term is absent because it is canceled by a homogeneous positive background introduced in the jellium model .
thus the surface tension energy in mebs takes a role similar to the homogenous positive background in jellium .
.several typical length and energy scales for the electron - ripplon system in the bubble are given in this table , for bubbles with different numbers of electrons and subjected to different pressures .
the first row lists the bubble radius in microns .
the second row lists the kinetic energy scale @xmath20 of electrons in the bubble , the energy of angular momentum level @xmath1 being @xmath32 .
the third row gives the electronic level splitting @xmath33 , at the fermi level , between the subsequent angular momentum levels of the spherical 2d electron gas .
the fourth row lists the energy scale for ripplons , @xmath34^{1/2}$ ] in @xmath35k ( 1 @xmath35k corresponds to 0.1309 mhz ) .
the fifth row gives the strength of the electric field at the bubble surface , pressing the electrons to the helium surface and resulting in electron - ripplon coupling .
the sixth row provides values for the electron - ripplon coupling constant @xmath36 in expression ( [ melrip ] ) . [ cols="<,^,^,^,^,^,^,^,^,^ " , ] * ripplons on the bubble surface * small - amplitude oscillations of the bubble surface can be quantized , leading to the concept of spherical ripplons .
the ripplon gas ( excluding the breathing mode ) is described by the hamiltonian @xmath37 the bare ripplonic frequencies for this system , at a pressure @xmath38 , are @xcite @xmath39 where @xmath40 j / m@xmath41 is the surface tension of helium , and @xmath42 kg / m@xmath43 is its density . in the surface tension dominated regime ( @xmath44 ) , @xmath45 with @xmath46^{1/2}$ ] .
typical ripplon frequencies lie in the mhz - ghz ( or @xmath35k ) range .
the ripplon green s function is defined by @xmath47\right\rangle \\ & \text{with } \hat{a}_{l , m}=\hat{a}_{l , m}+\hat{a}_{l ,- m}^{+},\end{aligned}\ ] ] where @xmath48 is the time ordering operator and @xmath49 is a sum of ripplon creation and annihilation operators .
the unperturbed ripplon propagator ( corresponding to a system described by @xmath50 above ) in the frequency domain is @xmath51 where @xmath52 is a positive infinitesimal . *
electron - ripplon interaction * the electron - ripplon interaction can be written as @xmath53 where @xmath54 is the electron - ripplon interaction amplitude and @xmath55 is the @xmath25 spherical component of the electron density .
the interaction between the electrons and the ripplons comes about due to the presence of an electric field , generated by the electrons themselves and pressing the electrons against the helium surface .
this is the electric pressing field @xmath56 , directed radially .
when a ripplon is present , it moves the electrons in the electric field generated by all other electrons and this results in an interaction energy .
the interaction energy is the product of the displacement caused by the ripplon and the electric field , summed for all electrons , similarly as in @xcite . rewriting this interaction energy in second quantization operators
, we find the interaction hamiltonian ( [ helrip ] ) with the interaction amplitude @xmath57^{1/4 } } , \label{melrip}%\ ] ] where the coupling constant due to the pressing electric field @xmath58 ( see table i ) is given by@xmath59 an additional contribution to the interaction energy between electrons and ripplons can be derived as the change in polarization energy of the electron - helium system when the helium surface is deformed and the electrons are at rest .
this mechanism for coupling was first derived by m. cole @xcite for electrons on a flat surface .
following the arguments of cole , we obtain a similar expression for the interaction amplitude ( [ melrip ] ) but with a different coupling constant @xmath60 where @xmath61 is the expectation value for the distance between the electron and the helium surface .
the total electron - ripplon coupling constant is then @xmath62 a single electron coupled to a bath of ripplons forms a ripplonic polaron @xcite . in a multielectron bubble
, the electric field pressing the electrons against the helium surface can be much larger than the field achievable on a flat helium surface , so that the ripplonic polarons will be in the strong coupling regime , and can even form a wigner lattice of ripplonic polarons @xcite . * cooper s argument * in this subsection , we follow cooper s argument for pairing @xcite and apply this to the present case of electrons and ripplons in the bubble .
the effective electron - electron interaction is the sum of the coulomb interaction between the electrons and a ripplon - mediated attractive interaction between the electrons . in @xcite
we drew a feynman diagram to represent the coulomb interaction between the electrons .
now , we can add another diagram with the same electron propagator lines , but instead of exchanging a virtual photon , exchanging a virtual ripplon .
this is shown in fig .
[ fig1 ] .
[ ptb ] figure1.eps the two diagrams in fig .
[ fig1 ] can be represented by a single diagram using an effective interaction @xmath63 both the coulomb and the ripplon - exchange interaction are given by a product of two vertex factors and one virtual particle propagator .
the total electron - electron interaction hamiltonian can be written as @xmath64 \\ & \times(-1)^{n}\hat{c}_{(l , m)\otimes(j , n),\sigma}^{+}\hat{c}_{(l^{\prime } , m^{\prime})\otimes(j ,-
n),\sigma^{\prime}}^{+}\hat{c}_{l^{\prime},m^{\prime } , \sigma^{\prime}}\hat{c}_{l , m,\sigma}.\end{aligned}\ ] ] let s study for which regimes @xmath65 is attractive , i.e. for which values of @xmath66 the ripplonic part dominates and is attractive .
it is clear that small energy transfers make the ripplonic part attractive because @xmath67 .
moreover , in general ripplon exchange will indeed occur with @xmath68 .
the reason for this is that the ripplonic energies are much smaller than the electronic level spacing , as can be seen from comparing rows 3 and 4 of table i. for ripplons with @xmath1 smaller than @xmath69 , the absorption or emission of a ripplon with angular momentum @xmath1 can not change the angular momentum of the electron due to energy conservation requirements .
the effective interaction at @xmath68 is @xmath70 the attractive interaction dominates strongly at small @xmath71 since @xmath72 as can be checked for typical bubbles by substituting the values from table i@xmath73 it is strongest for small @xmath1 and decreases roughly as @xmath74 : @xmath75 the strength of the effective interaction is @xmath76 .
since the @xmath36 is of the order of mk and @xmath77 of the order of @xmath35k ( see table i ) , the effective attractive interaction is large as compared the relevant energy scales @xmath36 , @xmath77 , and @xmath78 .
we are clearly in a strong coupling regime , in agreement with the results from @xcite .
* intralevel pairing * since the ripplon energies are much smaller than the electronic level spacing at the fermi level , the attractive interaction only takes place between two electrons on the same angular momentum level , and these electrons will be scattered into final states also on that angular momentum level .
an electron in angular momentum state @xmath79 , which emits a spherical ripplon in angular momentum state @xmath80 , finds itself in the following superposition of angular momentum states @xmath81 the projection of this final state @xmath82 on the angular momentum level @xmath1 of the initial state is @xmath83 & = \sqrt{\dfrac{2j+1}{4\pi}}\left\langle l,0;j,0|l,0\right\rangle \\ & \times\left\langle l , m;j ,-
n|l , m - n\right\rangle .\end{aligned}\ ] ] thus , the scattering process between two electrons with spin @xmath6 and @xmath84 on the angular momentum level @xmath1 can be described in second quantization as @xmath85f_{cg}[(l , m^{\prime}),(j , n)]\\ & \times v_{\text{eff}}(j , n;0)\hat{c}_{l , m - n,\sigma}^{+}\hat{c}_{l , m^{\prime } + n,\sigma^{\prime}}^{+}\hat{c}_{l , m^{\prime},\sigma^{\prime}}\hat { c}_{l , m,\sigma}%\end{aligned}\ ] ] this interaction hamiltonian derived for the multielectron bubble is already close to a bcs - like interaction hamiltonian .
it involves only electrons on the same angular momentum level and couples them with an attractive potential .
an initial state with a pair characterized by @xmath86 can be scattered into a pair with @xmath87 in various ways : namely by the scattering of a virtual ripplon with @xmath88 provided that @xmath89 is an even mode to obey the triangle rule of addition of angular momenta .
all these processes are indistinguishable ( initial and final states are exactly the same ) and their diagrams should be added to get the overall amplitude .
different combinations of clebsch - gordan coefficients will occur , which can take either a positive or a negative sign : the different diagrams can interfere constructively but also destructively .
the only case where we are sure that the different contributions will interfere constructively , is for pairs of electrons with opposite angular momentum @xmath90 .
the reason for this is that @xmath91 has the same sign as @xmath92 .
we investigated this point numerically , and found that the total effective interaction is indeed strongly reduced for pairs that do not have opposite angular momentum ( @xmath93 ) .
however , for @xmath94 we find that the effective interaction potential is not reduced and is only weakly dependent on @xmath95 . * effective hamiltonian * from the previous section we know that the interaction between the electrons can be written in the form of a bcs interaction hamiltonian given by @xmath96 where @xmath97 @xmath98 denotes spin up and spin down and @xmath99}^{2l}% \frac{2g^{2}}{\hbar\omega_{\sigma}}\frac{(j+1/2)}{(j+1)(j^{2}+j+2 + 2pr/\sigma ) } \nonumber\\ & \times f_{cg}[(l , m),(j , m^{\prime}-m)]f_{cg}[(l ,-
m),(j , m - m^{\prime } ) ] .
\label{vmm}%\end{aligned}\ ] ] due to the coefficients @xmath100 the summation can not run further than @xmath101 and only even values of @xmath89 contribute .
the summation starts from @xmath102 , or if this is less than 2 , it starts at @xmath103 .
the @xmath30 deformation is not taken into account ( it is the radius of the bubble ) , and the @xmath104 deformation is a uniform translation which can not couple to the internal degrees of freedom . to proceed , we will introduce an averaged interaction amplitude at the fermi angular momentum level : @xmath105 the interaction amplitude still depends on the angular momentum level @xmath1 .
but , as we shall see in the next section , pair correlations occur only in levels close to @xmath106 where the @xmath1-dependence of the interaction amplitude can be neglected .
so we use the value of the interaction amplitude at @xmath106 also for levels close to @xmath106 .
values for @xmath107 for various configurations are given in figure [ fig2 ] .
[ ptb ] figure2.eps with this , we have established that the properties of the ripplon - mediated electron - electron interaction lead to a cooper - type attractive interaction between the electrons , and to a bcs - like hamiltonian @xmath108 the peculiarity of this pairing hamiltonian is that the pairing takes place within discrete energy levels .
this effective hamiltonian for electrons pairing due to an attractive interaction brought about by ripplon exchange can be solved by introducing a variational many body wave function as in the bcs treatment .
however , we chose to apply the richardson method @xcite , initially developed in the context of nuclear physics and recently reintroduced @xcite to the condensed matter community to describe superconductivity in nanosize metallic grains .
this method is particularly suitable for finite systems with a discrete level structure such as the multielectron bubble .
having argued that multielectron bubbles are a suitable candidate to observe pairing of electrons in a spherical two - dimensional system , we will apply the richardson model to the effective pairing hamiltonian ( [ heff ] ) to gain insight in the properties of the paired phase .
note that the analysis here can be applied to the general problem of a s2deg with attractive interactions between electrons on the same angular momentum level .
the previous section provides possible values for the coupling constant @xmath107 ( see fig .
[ fig2 ] ) and for the relevant energy scales ( see table i ) , and a justification for the applicability of the effective pairing hamiltonian ( [ heff ] ) to multielectron bubbles specifically .
nevertheless the results derived in this section can be used to investigate other systems such as the superconducting properties of thin electronic nanoshells , or can be investigated as an academic question regarding spherical electronic systems .
the richardson model provides a method of solution for the so - called reduced bcs hamiltonian @xcite:@xmath109 now consider the hamiltonian ( [ heff ] ) that we derived in the previous section , and collect all terms that contain operators working on the angular momentum level @xmath1:@xmath110 note that only electrons within the same angular momentum energy level @xmath1 interact : only _ _
intra__level interactions take place , but no _ _ inter__level interactions .
this is due to the fact that the relevant ripplon energies are much smaller than the interlevel energy splitting ( see table i for typical values in multielectron bubbles ) .
so , the set of electrons with a given angular momentum @xmath1 can be considered as an independent subsystem , described by the hamiltonian ( [ hl ] ) .
the full system is the collection of independent subsystems characterized by different @xmath1 .
the full hamiltonian ( [ heff ] ) is just the sum of the hamiltonians ( [ hl ] ) describing independent subsystems with different @xmath1 : @xmath111 moreover , each of the hamiltonians ( [ hl ] ) corresponds to a reduced bcs hamiltonian ( [ hrich ] ) that can be solved with the richardson method .
indeed , setting @xmath112 and @xmath113 brings ( [ hl ] ) into the same form as ( [ hrich ] ) .
so , we have a collection of independent systems ( each characterized by a particular value of @xmath1 ) that can each be solved by the richardson method .
as shown by richardson @xcite , the exact solution of the reduced bcs hamiltonian for @xmath114 electron pairs amounts to solving a set of @xmath114 nonlinear coupled equations . in general , the aforementioned set of equations can be solved only by numerical computation .
however , in the particular case when all the involved single - particle states belong to one an the same energy level as is the case for a spherical multielectron bubble the energy of electron pairs can be easily found analytically ( see , e.g. , ref . ) .
the result for the energy of electrons in the subsystem characterized by angular momentum @xmath1 can be written down as @xmath115 the energy levels of the subsystem with angular momentum @xmath1 is characterized by three quantum numbers , @xmath116 , @xmath117 and @xmath118 . here
@xmath116 is the number of electrons pairs , @xmath117 is the number of unpaired electrons , and @xmath118 of elementary bosonic pair - hole excitations @xcite in the system of @xmath116 pairs . to better understand these quantum numbers , consider two bare single - electron states @xmath119 and @xmath120 .
if both states are occupied , this represents an electron pair .
the electron pair can scatter into another pair of states @xmath121 under the influence of the interaction term in the hamiltonian ( [ hl ] ) .
the number of such pairs with given @xmath1 is @xmath122 and it has to be less than or equal to @xmath2 .
now consider the case where only one state of the pair @xmath119 and @xmath123 is occupied .
then we have an unpaired electron that can not participate in the scattering described by the interaction term in ( [ hl ] ) . moreover
, electron pairs can not scatter into the pair of states @xmath119,@xmath120 because one of these states is already occupied .
the states @xmath119,@xmath120 are then blocked for scattering of pairs .
the number of these blocked spin - degenerate bare states equals the number of unpaired electrons , and is denoted by @xmath117 .
the total number of electrons in angular momentum level @xmath1 is then @xmath124and this has to be less than or equal to @xmath125 .
the last quantum number @xmath118 indexes the so - called ` pair - hole excitations ' @xcite .
these are bosonic excitations that involve a redistribution of the amplitude for correlated pairs over the @xmath2 states with angular momentum @xmath1 .
note that @xmath126 corresponds to the ground state for the correlated pairs .
the total energy can be expressed as a sum of the energies for each independent subsystem : @xmath127 the state of the entire system is characterized by a large set of quantum numbers , three ( @xmath128 ) for each angular momentum subsystem . how do we characterize the ground state of electrons in a spherical bubble with pairing interactions ? in bulk bcs superconductors only electrons in an energy band of the debye energy @xmath129 around the fermi level participate in the pairing . however , in the present case , the relevant ripplon frequencies are much smaller than the splitting between consecutive @xmath1 levels near the fermi energy . therefore one might argue that pairing correlations only take place in the one subsystem with @xmath28 where @xmath130 is the angular momentum at the fermi level . indeed , all levels above the fermi level ( @xmath131 ) are empty , and all levels below the fermi level ( @xmath132 ) are completely filled so that no pairing correlations can be achieved by ripplon - mediated scattering .
yet this turns out to be wrong .
the main difference between pairing in the present case and pairing in conventional superconductors is that in conventional bcs superconductors the gap @xmath133 is much smaller than the relevant phonon energy ( @xmath134 ) , whereas in the present case the interaction energy per electron can be larger than the relevant ripplon energies ( @xmath135 ) and even larger than the level splitting ( @xmath136 ) .
this can be inferred by comparing the interaction strengths listed in table ii to the ripplon energies and level splittings reported in table i. it then becomes energetically advantageous to redistribute the electrons amongst the energy levels @xmath1 .
indeed , by promoting a pair of electrons from a level @xmath137 below the fermi level @xmath138 to a level @xmath139 above the fermi level ( @xmath140 ) , both these levels @xmath141 can also form pairing correlations .
the energy gain by forming pairing correlations is larger than the energy needed to promote the electrons form level @xmath137 to level @xmath142 .
so we obtain the remarkable result that for sufficiently large interaction strength @xmath107 ( as we think is the case for mebs ) , although only intralevel scattering can take place , still levels well below and above the fermi level can be affected . to see this in more detail ,
let us consider for example a bubble with an even number of electrons .
the ground state in an even bubble is achieved by @xmath126 ( no pair - hole excitations ) and @xmath143 ( no unpaired electrons ) . in order to find the ground - state configuration @xmath116 , consider the change in the total energy due to a transfer of an electron pair from the @xmath144 level to the next higher level .
( [ el ] ) with @xmath126 and @xmath143 , we find@xmath145 where @xmath116 and @xmath146 are the number of electron pairs on the levels @xmath147 and @xmath148 , respectively , before the transfer .
the first term in the right - hand side of ( [ detransfer ] ) represents the energy cost in promoting a pair of electrons from level @xmath1 to level @xmath148 , and the second term corresponds to the gain in energy due to pairing correlations .
as seen from ( [ detransfer ] ) , such a transfer reduces the total energy if the inequality @xmath149 is satisfied .
this inequality is not satisfied if @xmath150 , so that an inverted population of the levels obviously never appears in the ground state .
condition ( [ condtransf ] ) is also not fulfilled at weak interaction , i.e. for @xmath151 . in the case of @xmath152
the above definition of weak interaction can be simplified to @xmath153 . from table
i and table ii , we infer that for mebs it is likely that @xmath154 so that a redistribution of electron pairs as described in the previous paragraph is indeed energetically favorable . fig .
[ fig3 ] illustrates the ground - state configuration obtained by minimizing the total energy ( [ etotal ] ) with respect to the @xmath116 s ( and setting @xmath155 ) .
the results are shown as a function of @xmath156 for fixed @xmath157 .
the left panel shows the result for @xmath158 electrons on the fermi level in the ground state of an meb with @xmath159 this corresponds to roughly half - filling of the fermi level .
the right panel shows the result for @xmath160 , a closed - shell configuration in the @xmath161 ground state .
blue ( green ) color corresponds to filled ( empty ) states on the energy levels .
switching off the interactions , @xmath161 , we see that all levels below @xmath157 are completely filled ( blue color ) , and all levels above @xmath106 are completely empty ( green color ) . for @xmath162 , some empty states ( green )
appear in levels @xmath163 that are below @xmath106 , and some electron pairs appear in levels @xmath164 that are above @xmath106 .
note that @xmath107 needs to exceed a critical value ( of the order of @xmath165 ) for the redistribution of electron pairs to take place .
also note that approximately @xmath166 levels around the fermi level are affected by the redistribution of electron pairs .
[ ptb ] figure3.eps having obtained , for the ground state , the total energy @xmath167 , we can derive the condensation energy @xmath168 where @xmath169 [ @xmath170 is the ground - state energy in the presence [ in the absence ] of the pairing interaction .
the last term in the rhs describes the interaction energy for the uncorrelated fermi ground state and @xmath171 means the integer part of @xmath172 . in fig .
[ fig4 ] we plot the condensation energy @xmath173 as a function of @xmath156 .
first consider the regime of strong interactions , @xmath174 .
for this regime , we find that the @xmath175dependence of the condensation energy becomes negligible .
the condensation energy rapidly rises with @xmath156 , approximately as @xmath176 .
next , consider the regime of weak interactions ( @xmath177 ) where , as we have seen , the electron redistribution between the energy levels is absent . in this case
the condensation energy is a linear function of @xmath107 , and such a linear behavior can indeed be seen in fig .
[ fig4 ] at @xmath177 . in the regime of weak interaction ,
the condensation energy is strongly influenced by @xmath178 : it is zero for closed shell configurations and reaches a maximum at half filling . for typical mebs , we find that @xmath179 is of the order of several kelvin .
decreasing the number of electrons or pressurizing the bubble increases the condensation energy .
bigger bubbles have a smaller condensation energy .
[ ptb ] figure4.eps how can the pairing correlations be probed experimentally ?
what properties will distinguish the state with pair correlations from the normal fermi sea ? in the preceding subsection , we have found the ground state for the system described by the hamiltonian @xmath180 , expression ( [ heff ] ) , and in the preceding section we argued , based on cooper s argument , why this hamiltonian is suitable to describe mebs .
nevertheless , more exotic correlated or magnetic states not described by the hamiltonian ( [ heff ] ) can not be completely excluded as candidates for the true ground state : only experiment will give a decisive answer as to what the realized state in the meb will be .
it is therefore very relevant to discuss accessible experimental signatures of the correlated many - body state that we have described
. in metallic nanograins and nanowires , pairing correlations ( leading to superconductivity ) were revealed through a measurement of the density of states @xcite . also here , we propose to reveal pairing correlations by probing the density of states . in the case of stabilized mebs @xcite , the density of states can be measured by spectroscopy . also tunneling experiments
are possible by placing an electrode close to the bubble , and reducing the thickness of the layer of helium between the bubble and the helium .
the density of states that would be revealed by these experiments is the subject of the next subsection . in the previous subsection
, we found that the ground state is characterized by a redistribution of electrons between degenerate bare energy levels , as compared to the ground - state electron distribution in the absence of interaction .
here we analyze the effect of the pairing interaction on the density of ( many - electron ) states in mebs .
this density of states can be written down in general as @xmath181 from expression ( [ etotal ] ) it is clear that the many - electron energy levels @xmath182 are characterized by a set of quantum numbers @xmath183 @xmath184 .
the summation runs over all possible sets @xmath183 of quantum numbers .
the degeneracy of the many - electron energy level @xmath182 is denoted by @xmath185 .
consider one particular angular momentum state @xmath1 with a given \{@xmath128}. the number of ways to place the unpaired electrons , multiplied by the number of pair - hole excitations with given @xmath118 , is @xmath186 ( see refs .
@xcite ) : @xmath187{l}% 1,\quad g_{l}=0\\ \left ( c_{g_{l}}^{2l+1-b_{l}}-c_{g_{l}-1}^{2l+1-b_{l}}\right ) , \quad g_{l}\geq1 \end{array } \right.\ ] ] with @xmath188 , the binomial coefficients .
the total degeneracy is the product of the degeneracies of the independent systems , @xmath189 for graphical representation , it is more convenient to consider instead of @xmath190 the quantity @xmath191 which gives the number of ( many - electron ) states in the energy range of width @xmath192 around the energy @xmath58 . fig .
[ fig5 ] gives an example of the calculated @xmath193 .
the calculations are performed for @xmath194 electrons in the fermi angular momentum level @xmath157 , so that in the ground state at @xmath161 the fermi level is approximately half filled .
together with the whole spectrum of excitations ( displayed in light green ) we also show ( in dark blue ) the intralevel excitations from the ground state .
[ ptb ] figure5.eps at @xmath161 , as shown in fig .
[ fig5]a , there is a set of excitations that correspond to different interlevel electron transitions between the single - electron bare energy levels with @xmath1 close to @xmath106 ( obviously , intralevel transitions have zero energy in this case ) . at @xmath161 ,
all the many - electron energy levels in the system under consideration are highly degenerate .
this is due to the numerous possibilities to distribute electrons over single - electron states belonging to a partially filled energy level with a given @xmath1 . at small nonzero @xmath107 , the many - electron energy levels for each angular momentum level are split .
excitations involving pair breaking @xmath195 and pair - hole excitations @xmath196 start to affect the density of states .
the result , shown in fig .
[ fig5]b , is the appearance of energy bands in the density of states .
the first band corresponds to intralevel excitations from the ground state , the other bands involve interlevel excitations from one angular momentum state to another . at this point
it is interesting to investigate the gap in the excitation spectrum for intralevel excitations . in superconducting nanograins ,
measuring such a spectroscopic gap signals the onset of superconductivity @xcite .
first , consider a process where the ground state ( with @xmath116 pairs and @xmath143 unpaired electrons ) is transformed into a final state with @xmath197 and @xmath198 unpaired electrons .
we will refer to such a process as a ` pair - breaking excitation ' .
a pair - breaking ` gap ' @xmath199 can then be defined as the energy necessary to create a pair - breaking excitation from the ground state:@xmath200 using ( [ el ] ) , @xmath201 similarly , we can calculate the smallest energy needed to create a pair - hole excitation ( @xmath202 ) and define a pair - hole excitation gap @xmath203@xmath204
so , in the system under consideration , the smallest energy for a pair - breaking excitation equals that for a pair - hole excitation , @xmath205 , and we can drop the superscripts ( p - b ) and ( p - h ) . at @xmath206
the intralevel excitation gap @xmath133 approaches the energy spacing @xmath207 between single - electron bare energy levels with consecutive values of @xmath1 .
this means that , for @xmath208 , interlevel excitations exist with energy smaller than the gap @xmath209 moreover , as implied by eq .
( [ detransfer ] ) , an increase of @xmath107 can substantially reduce the energies of interlevel transitions of pairs . at @xmath210
( see fig .
[ fig5]c ) , the lowest interlevel excitations ( green ) already have energies smaller than @xmath211(first nonzero blue line ) . at even larger @xmath107
, pairing correlations appear on many different angular momentum levels .
the pair - breaking / pair - hole excitation gap , @xmath212 , depends on the angular momentum .
therefore , as seen from fig .
[ fig4]d , the peaks of @xmath193 corresponding to intralevel excitations from the ground state , are split into peaks corresponding to different angular momentum states where pairing can take place . with increasing @xmath107 , the excitation spectrum tends to become quasi - continuous , with jumps of several orders of magnitude in @xmath193 near the energies corresponding to pair - breaking or pair - hole excitations . between these jumps
there is a more uniform distribution of excitations , corresponding to interlevel transitions of pairs .
[ fig6 ] provides an overview of the behavior of @xmath193 as a function of both @xmath58 and @xmath107 for fixed @xmath157 and two different values of @xmath213 .
one can see an interplay between intralevel excitations , whose energies always increase with increasing @xmath107 , and excitations corresponding to interlevel transitions of pairs . for the latter , both an increase and a decrease in energy are possible with increasing @xmath107 . while at @xmath214 the patterns of @xmath215 are very different for different @xmath213 , at larger @xmath107 the behavior of @xmath193 in mebs with a definite parity of the number of electrons becomes almost independent of the precise value of this number [ cf
[ fig6](a ) and [ fig6](b ) ] .
this universal behavior of @xmath216 at large @xmath156 is further illustrated by fig .
[ fig6 ] , where the calculated dependence of @xmath217 on @xmath218 is shown for different @xmath213 and @xmath106 . as seen from fig .
[ fig6 ] , neither moderate changes of @xmath106 nor variations of @xmath213 at a fixed @xmath106 significantly affect the shape of @xmath217 versus @xmath218 for mebs with a given parity of the number of electrons . at the same time , fig .
[ fig6 ] demonstrates a pronounced difference between the results for even mebs and those for odd mebs .
[ ptb ] fig6small.eps [ ptb ] figure7.eps summarizing this subsection , we find that in the regime most relevant for multielectron bubbles , namely @xmath219 , the pairing correlations reveal themselves in the density of states not as a spectroscopic gap , but rather as a significant jump in the density of states at the pair - breaking energy @xmath133 .
moreover , in this regime ( @xmath219 ) the density of states is less sensitive to even - odd effects .
the observation ( through tunneling or spectroscopy ) of a jump in the density of states at @xmath133 can be used as a way to infer the presence of pairing correlations in the meb . moreover
, the pair - breaking energy can be used to estimate a temperature @xmath220 above which the pairing correlations will be suppressed : when @xmath221 the thermal energy is large enough to support an appreciable amount of pair - breaking excitations . for typical mebs ,
this temperature is of the order of kelvins .
smaller bubbles or compressed bubbles have a larger gap and thus a larger @xmath220 .
in the first part of this paper , we have analyzed how the electron - ripplon interaction on a spherical surface may lead to an attractive effective electron - electron interaction and give rise to a cooper pairing scenario .
the effective hamiltonian of the two dimensional spherical electron system is mapped on a bcs - type hamiltonian and typical values of energies , length scales and interaction strengths are estimated . in the second part we use richardson s method to investigate pairing properties of a two dimensional spherical electron system .
we find that when the condensation energy per pair is larger than the bubble energy scale @xmath222 the ground state of the system acquires unique properties that set it apart from pairing in conventional superconductors or superconducting nanograins .
in particular , we show that although only intralevel interactions are included ( since the relevant ripplon energies are smaller than the level splitting ) , electron pairs nevertheless redistribute themselves among the different levels and pairing takes place in an interval of energies around the fermi energy , much larger than the typical ripplon energy .
the density of states reveals an intricate interplay between intralevel transitions and interlevel excitation of pairs , evolving from a discrete spectrum typical for confined systems to a staircase quasi - continuum upon increasing interaction strength . at strong coupling
the density of states reveal the presence of pairing correlations not through a spectroscopic gap as in metallic nanograins @xcite , but through a jump in the density of states at the pair - breaking energy .
these results show that spherical electron systems reveal particularly interesting pairing properties , distinct from their bulk or flat - surface counterparts , combining both topological effects and confinement effects .
mebs are a particularly pure realization of the two - dimensional electron system ( just as an electron film on helium forms a pure realization of a flat 2deg )
. thin nanoshells ( monolayer gold coatings of a non - conducting nanograin ) may be another realization of this system .
is supported financially by the fonds voor wetenschappelijk onderzoek - vlaanderen .
this research has been supported financially by the fwo - v projects nos .
g.0435.03 , g.0306.00 , g.0274.01n , the w.o.g .
project wo.035.04n , the goa bof ua 2000 . also financial support from the department of energy , grant no .
de - fg02-er45978 is acknowledged .
j.t . gratefully acknowledges support of the special research fund of the university of antwerp , bof noi ua 2004 .
r. d. averitt , d. sarkar , and n. j. halas , phys . rev . lett . * 78 * , 4217 ( 1997 ) ; s. j. oldenburg , r. d. averitt , s. l. westcott , and n. j. halas , chem . phys
. lett . * 288 * , 243 ( 1998 ) ; e. prodan , p. nordlander , and n. j. halas , chem .
. lett . * 368 * , 94 ( 2003 ) .
i. f. silvera , j. tempere , j. huang , j. t. devreese , in : _ frontiers of high - pressure research ii : application of high pressure to low - dimensional novel electronic materials _
h. d. hochheimer et al . ,
kluwer academic publ .
2001 ) , p. 541 - 549 . | electrons in a multielectron bubble in helium form a spherical , two - dimensional system coupled to the ripplons at the bubble surface . the electron - ripplon coupling ,
known to lead to polaronic effects , is shown to give rise also to cooper pairing . a bardeen - cooper - schrieffer ( bcs )
hamiltonian arises from the analysis of the electron - ripplon interaction in the bubble , and values of the coupling strength are obtained for different bubble configurations .
the bcs hamiltonian on the sphere is analysed using the richardson method .
we find that although the typical ripplon energies are smaller than the splitting between electronic levels , a redistribution of the electron density over the electronic levels is energetically favourable as pairing correlations can be enhanced .
the density of states of the system with pairing correlations is derived .
no gap is present , but the density of states reveals a strong step - like increase at the pair - breaking energy .
this feature of the density of states should enable the unambiguous detection of the proposed state with pairing correlations in the bubble , through either capacitance spectroscopy or tunneling experiments , and allow to map out the phase diagram of the electronic system in the bubble . |
since the seminal work of bachelier @xcite , prices in a competitive market have been modeled as a memoryless stochastic process .
in fact , according to the efficient market hypothesis ( emh ) , prices fully reflect all available information @xcite .
this property was duly proved by samuelson @xcite .
it is known that informational efficiency can vary over time .
this may be due to several reasons .
for example , cajueiro and tabak @xcite studied the chinese stock market and found that liquidity plays a role in explaining the evolution of the long term memory ; bariviera @xcite argued that informational efficiency in the thai stock market is influenced by liquidity constraints ; while bariviera _
@xcite detailed the influence of the 2008 financial crisis on the memory endowment of the european fixed income market .
however , changes in informational efficiency are essentially unpredictable , under the emh framework .
the problem addressed in this paper stems from a newspaper .
mollenkamp and whitehouse @xcite published a disruptive article in the wall street journal : they suggested that the libor rate did not reflect what it was expected , _
i.e. _ , the cost of funding of prime banks . the british bankers association ( bba ) defines libor as `` ... the rate at which an individual contributor panel bank could borrow funds , were it to do so by asking for and then accepting inter - bank offers in reasonable market size , just prior to 11:00 [ a.m. ] london time '' .
every london business day , each bank in the contributor panel ( selected banks from bba ) makes a blind submission such that each banker does not know the quotes of the other bankers . a compiler , thomson reuters , then averages the second and third quartiles .
this average is published and represents the libor rate on a given day .
in other words , _ libor is a trimmed average of the expected borrowing rates of leading banks . _
libor rates has been published for ten currencies and fifteen maturities .
as it is defined , libor is expected to be the best self estimate of leading banks borrowing cost at different maturities .
several publications in newspapers @xcite casting doubts on libor integrity triggered investigations by several surveillance authorities such as the us department of justice , the uk financial services authority or the european commission .
all these offices found several traces of misconduct , which resulted in severe fines to leading banks .
there are a few papers dealing with this topic in academic journals .
most of them are focused , once the manipulation is known , on verifying or discarding its existence , by means of statistical tests .
taylor and williams @xcite documented the detachment of the libor rate from other market rates such as overnight interest swap ( ois ) , effective federal fund ( eff ) , certificate of deposits ( cds ) , credit default swaps ( cds ) , and repo rates .
snider and youle @xcite studied individual quotes in the libor bank panel and found that libor quotes in the us were not strongly related to other bank borrowing cost proxies .
abrantes - metz _ et al . _
@xcite analyzed the distribution of the second digits ( sds ) of daily libor rates between 1987 and 2008 and , compared it with uniform and benford s distributions .
if we take into account the whole period , the null hypothesis that the empirical distribution follows either the uniform or benford s distribution can not be rejected .
however , if we take into account only the period after the subprime crisis , the null hypothesis is rejected .
this result calls into question the aseptic " setting of libor .
monticini and thornton @xcite found evidence of libor under - reporting after analyzing the spread between 1-month and 3-month libor and the rate of certificate of deposits using the bai and perron @xcite test for multiple structural breaks . for a historical overview of the libor case from a regulator point of view , see the federal reserve staff report @xcite .
recently , bariviera _ et al . _
@xcite present preliminary results about 3-month uk libor manipulation .
they performed a symbolic time series analysis using the complexity entropy causality plane ( cecp ) .
our approach goes deeper in two aspects .
first , we study the behavior of the libor for several maturities and currencies .
second , we introduce a local information theory quantifier , which is able to detect tiny perturbations in the probability density function .
consequently , instead of analyzing our results with global vs. global quantifiers , as in the cecp , we study the time series by means of global vs. local quantifiers , defined in the shannon - fisher plane .
interest rates have broad and wide economic impact on our daily life . the lack of integrity of libor as an information signal gives market participants a wrong proxy of borrowing costs , thus providing a bad rate for pricing financial products .
many mortgages as well as sovereign bonds have libor - linked interest rates .
therefore , the wrong setting of the interest rates produces a cascade effect throughout the whole economy .
consequently , libor reliability is crucial to private and public borrowers around the world .
our approach is somewhat different to the previous literature .
the aim of this paper is to propose a quantitative technique based on information theory quantifiers , in order to detect changes in the stochastic / chaotic underlying dynamics of the libor time series .
we claim that our methodology is able to capture changes in the process dynamics on an ongoing basis . in other words ,
our method allows periodic monitoring of interest time series .
this market audit , done at regular intervals , can detect unusual mutation in the statistical properties of a time series .
it is clear that not only manipulation can shift signal features .
other external forces , such as increasing noise or special economic situations influence on signal observation and measurement .
therefore , results should be read with care . a change in the stochastic dynamics
should not be read exclusively as produced by manipulation .
it could be due to spurious contamination of the signal , or liquidity constraints , etc
. however , our method acts as an early warning mechanism to detect some kind of market distress " .
this study is relevant no only for researchers , but also for surveillance authorities who need an efficient market watch device in order to detect strange movements in key economic variables such as the libor .
this paper is structured as follows .
section [ sec : methodology ] describes the methodology .
section [ sec : data ] details the data used in this paper .
section [ sec : results ] discuses our empirical findings .
finally , section [ sec : conclusions ] draws the main conclusions of our research .
financial markets are complex , dynamic systems in which unobservable hidden structures govern their behavior .
usually , we can only observe an output , _
e.g. _ an equilibrium price , an interest rate fixing , etc . as a consequence
, the researcher should study the behavior of that output in order to infer the characteristics of the underlying dynamical phenomenon .
time series should be carefully analyzed in order to extract relevant information for simulation and forecasting purposes .
information theory derived quantifiers can be considered good candidates for this task because they are able to characterize some properties of the probability distribution associated with the observable or measurable quantity .
if the libor rates were somewhat manipulated ( as suggested in ref .
@xcite ) , some change in the stochastic behavior should appear . to be more precise , according to the emh , interest rate time series should behave approximately as an standard brownian motion .
manipulation is , by definition , the introduction of an strange deterministic device into the time series .
according to wold @xcite a time series can be split into two components , one purely random and other deterministic .
if manipulation is successful , the deterministic induced behavior should offset the random behavior .
this process results in an reduction of the natural stochastic character of the interest rates .
we argue that the selected information theory quantifiers are able to detect such reduction and its temporal duration .
we use two specific quantifiers : permutation shannon entropy and permutation fisher information measure .
these quantifiers are evaluated in pairs , displaying them in @xmath0-planar representation . this causal shannon
fisher plane gives an insight on global versus local perturbations on the behavior of a time series of the underlying dynamics of a physical process under analysis .
the shannon entropy is usually regarded as a natural measure of the quantity of information in a physical process .
given a continuous probability distribution function ( pdf ) @xmath1 with @xmath2 and @xmath3 , its associated _ shannon entropy _
@xmath4 @xcite is @xmath5~=~-\int_{\delta}~f~\ln(f)~dx \ , \ ] ] a measure of global character " that it is not too sensitive to strong changes in the distribution taking place on a small - sized region .
let now @xmath6 be a discrete probability distribution , with @xmath7 the number of possible states of the system under study . in the discrete case
, we define a normalized " shannon entropy , @xmath8 , as @xmath9~=~ s[p ] / s_{max } ~=~\left\{-\sum_{i=1}^{n}~p_i~\ln ( p_i ) \right\ } / s_{max } \ , \ ] ] where the denominator @xmath10 = \ln n$ ] is that attained by a uniform probability distribution @xmath11 .
the _ fisher s information measure _ ( fim ) @xmath12 @xcite , constitutes a measure of the gradient content of the distribution @xmath1 , thus being quite sensitive even to tiny localized perturbations .
it reads @xmath13~=~\int_{\delta}~ { { 1 } \over { f(x ) } } \left [ { { df(x ) } \over { dx } } \right]^2 ~dx ~=~4 \int_{\delta}~\left [ { { d \psi(x ) } \over { dx } } \right]^2 \ .\ ] ] fim can be variously interpreted as a measure of the ability to estimate a parameter , as the amount of information that can be extracted from a set of measurements , and also as a measure of the state of disorder of a system or phenomenon @xcite . in the previous definition of fim ( eq .
( [ fisher ] ) ) the division by @xmath1 is not convenient if @xmath14 at certain @xmath15values .
we avoid this if we work with a real probability amplitudes @xmath16 @xcite , which is a simpler form ( no divisors ) and shows that @xmath12 simply measures the gradient content in @xmath17 .
the gradient operator significantly influences the contribution of minute local @xmath18variations to fim s value .
accordingly , this quantifier is called a local " one @xcite .
let @xmath6 be a discrete probability distribution , with @xmath7 the number of possible states of the system under study .
the concomitant problem of information - loss due to discretization has been thoroughly studied and , in particular , it entails the loss of fim s shift - invariance , which is of no importance for our present purposes @xcite . for the fim
we take the expression in term of real probability amplitudes as starting point , then a discrete normalized fim , @xmath19 , convenient for our present purposes , is given by @xmath20~=~f_0~\sum_{i=1}^{n-1}~[(p_{i+1})^{1/2 } - ( p_{i})^{1/2}]^2 \ .\ ] ] it has been extensively discussed that this discretization is the best behaved in a discrete environment @xcite .
here the normalization constant @xmath21 reads @xmath22 if our system lies in a very ordered state , which occurs when almost all the @xmath23 values are zeros except for a particular state @xmath24 with @xmath25 , we have a normalized shannon entropy @xmath26 and a normalized fisher s information measure @xmath27 . on the other hand , when the system under study is represented by a very disordered state , that is when all the @xmath23 values oscillate around the same value we obtain @xmath28 while @xmath29 .
one can state that the general fim behavior of the present discrete version ( eq .
( [ fisher - disc ] ) ) , is opposite to that of the shannon entropy , except for periodic motions @xcite .
the local sensitivity of fim for discrete pdfs is reflected in the fact that the specific @xmath30ordering " of the discrete values @xmath23 must be seriously taken into account in evaluating the sum in eq .
( [ fisher - disc ] ) .
this point was extensively discussed by rosso _
_ in previous works @xcite .
the summands can be regarded as a kind of distance " between two contiguous probabilities .
thus , a different ordering of the pertinent summands would lead to a different fim - value , hereby its local nature . in the present work ,
we follow the lexicographic order described by lehmer @xcite in the generation of bandt - pompe pdf ( see next section ) . given the local character of fim , when combined with a global quantifier as the shannon entropy , conforms the shannon fisher plane , @xmath31 , introduced by vignat and bercher @xcite .
these authors showed that this plane is able to characterize the non - stationary behavior of a complex signal .
time series ( ts ) analysis , _
i.e. _ temporal measurements of variables , is a very important area of economic science . in particular
, it is very important to extract information in order to unveil the underlying dynamics of irregular and apparently unpredictable behavior in a given market .
the starting point of ts analysis is to determine the most appropriate probability density function associated with the ts .
several methods compete for its proper estimation .
for example rosso _
_ @xcite propose the frequency of occurrence ; de micco _ et al .
_ @xcite propose amplitude - based procedures ; mischaikow _ et al . _
@xcite prefer binary symbolic dynamics ; powell _ et al .
_ @xcite recommend fourier analysis and ; rosso _ et al . _
@xcite introduce wavelets transformation , among others .
the suitability of each of the proposed methodologies depends on the peculiarity of data , such as stationarity , length of the series , the variation of the parameters , the level of noise contamination , etc .
in all these cases , global aspects of the dynamics can be somehow captured , but the different approaches are not equivalent in their ability to discern all relevant physical details .
bandt and pompe @xcite introduced a simple and robust symbolic method that takes into account time order of the ts .
the pertinent symbolic data are : _
( i ) _ created by ranking the values of the series ; and _ ( ii ) _ defined by reordering the embedded data in ascending order , which is tantamount to a phase space reconstruction with embedding dimension ( pattern length ) @xmath32 and time lag @xmath33 . in this way , it is possible to quantify the diversity of the ordering symbols ( patterns ) derived from a scalar time series .
note that the appropriate symbol sequence arises naturally from the time series , and no model - based assumptions are needed .
in fact , the necessary `` partitions '' are devised by comparing the order of neighboring relative values rather than by apportioning amplitudes according to different levels .
this technique , as opposed to most of those in current practice , takes into account the temporal structure of the time series generated by the physical process under study .
this feature allows us to uncover important details concerning the ordinal structure of the time series @xcite and can also yield information about temporal correlation @xcite .
it is clear that this type of analysis of a time series entails losing some details of the original series amplitude information .
nevertheless , by just referring to the series intrinsic structure , a meaningful difficulty reduction has indeed been achieved by bandt and pompe with regard to the description of complex systems .
the symbolic representation of time series by recourse to a comparison of consecutive ( @xmath34 ) or nonconsecutive ( @xmath35 ) values allows for an accurate empirical reconstruction of the underlying phase - space , even in the presence of weak ( observational and dynamic ) noise @xcite .
furthermore , the ordinal patterns associated with the pdf is invariant with respect to nonlinear monotonous transformations . accordingly ,
nonlinear drifts or scaling artificially introduced by a measurement device will not modify the estimation of quantifiers , a nice property if one deals with experimental data ( see , _ e.g. _ , @xcite ) .
these advantages make the bandt and pompe methodology more convenient than conventional methods based on range partitioning ( _ i.e_. , pdf based on histograms ) .
to use the bandt and pompe @xcite methodology for evaluating the pdf , @xmath36 , associated with the time series ( dynamical system ) under study , one starts by considering partitions of the pertinent @xmath32-dimensional space that will hopefully `` reveal '' relevant details of the ordinal structure of a given one - dimensional time series @xmath37 with embedding dimension @xmath38 ( @xmath39 ) and embedding time delay @xmath33 ( @xmath40 ) .
we are interested in `` ordinal patterns '' of order ( length ) @xmath32 generated by @xmath41 which assigns to each time @xmath42 the @xmath32-dimensional vector of values at times @xmath43 .
clearly , the greater the @xmath44value , the more information on the past is incorporated into our vectors . by `` ordinal pattern '' related to the time @xmath45 ,
we mean the permutation @xmath46 of @xmath47 $ ] defined by @xmath48 in order to get a unique result , we set @xmath49 if @xmath50 .
this is justified if the values of @xmath51 have a continuous distribution , so that equal values are very unusual .
for all the @xmath52 possible orderings ( permutations ) @xmath53 when embedding dimension is @xmath32 , their associated relative frequencies can be naturally computed according to the number of times this particular order sequence is found in the time series , divided by the total number of sequences , @xmath54 in the last expression the symbol @xmath55 stands for number " .
thus , an ordinal pattern probability distribution @xmath56 is obtained from the time series .
consequently , it is possible to quantify the diversity of the ordering symbols ( patterns of length @xmath32 ) derived from a scalar time series , by evaluating the so - called permutation entropy ( shannon entropy ) and permutation fisher information measure .
of course , the embedding dimension @xmath32 plays an important role in the evaluation of the appropriate probability distribution , because @xmath32 determines the number of accessible states @xmath52 and also conditions the minimum acceptable length @xmath57 of the time series that one needs in order to work with reliable statistics @xcite . regarding the selection of the parameters , bandt and pompe suggested working with @xmath58 and specifically considered an embedding delay @xmath34 in their cornerstone paper @xcite .
nevertheless , it is clear that other values of @xmath33 could provide additional information .
it has been recently shown that this parameter is strongly related , if it is relevant , to the intrinsic time scales of the system under analysis @xcite .
additional advantages of the method reside in _
i ) _ its simplicity ( we need few parameters : the pattern length / embedding dimension @xmath32 and the embedding delay @xmath33 ) and _ ii ) _ the extremely fast nature of the pertinent calculation - process @xcite . the bp methodology can be applied not only to time series representative of low dimensional dynamical systems but also to any type of time series ( regular , chaotic , noisy , or reality based ) . in fact , the existence of an attractor in the @xmath32-dimensional phase space in not assumed . the only condition for the applicability of the bp method is a very weak stationary assumption : for @xmath59 , the probability for @xmath60 should not depend on @xmath61 . for a review of bp s methodology and its applications to physics ,
biomedical and econophysics signals see zanin _ et al .
rosso @xcite show that the above mentioned quantifiers produce better descriptions of the process associated dynamics when the pdf is computed using bp rather than using the usual histogram methodology .
the bandt and pompe proposal for associating probability distributions to time series ( of an underlying symbolic nature ) constitutes a significant advance in the study of nonlinear dynamical systems @xcite .
the method provides univocal prescription for ordinary , global entropic quantifiers of the shannon - kind .
however , as was shown by rosso and coworkers @xcite , ambiguities arise in applying the bandt and pompe technique with reference to the permutation of ordinal patterns .
this happens if one wishes to employ the bp - probability density to construct local entropic quantifiers , like the fisher information measure , which would characterize time series generated by nonlinear dynamical systems .
the local sensitivity of the fisher information measure for discrete pdfs is reflected in the fact that the specific `` @xmath62-ordering '' of the discrete values @xmath63 must be seriously taken into account in evaluating the sum in eq .
( [ fisher - disc ] ) .
the pertinent numerator can be regarded as a kind of `` distance '' between two contiguous probabilities .
thus , a different ordering of the pertinent summands would lead to a different fisher information value .
in fact , if we have a discrete pdf given by @xmath64 , we will have @xmath65 possibilities for the @xmath62-ordering .
the question is , which is the arrangement that one could regard as the `` proper '' ordering ?
the answer is straightforward in some cases , the histogram - based pdf constituting a conspicuous example .
for such a procedure , one first divides the interval @xmath66 $ ] ( with @xmath67 and @xmath68 the minimum and maximum amplitude values in the time series ) into a finite number on non - overlapping sub - intervals ( bins ) .
thus , the division procedure of the interval @xmath66 $ ] provides the natural order sequence for the evaluation of the pdf gradient involved in the fisher information measure . in our current paper , we chosen for the bandt
pompe pdf the lexicographic ordering given by the algorithm of lehmer @xcite , amongst other possibilities , due to it provide a better distinction of different dynamics in the shannon fisher plane , @xmath31 ( see @xcite ) .
we analize the libor rates in british pounds ( gbp ) , euro ( eur ) , swiss franc ( chf ) and japanese yen ( jpy ) , for the following seven maturities : overnight ( o / n ) , one week ( 1w ) , one month ( 1 m ) , two months ( 2 m ) , three months ( 3 m ) , six months ( 6 m ) and twelve months ( 12 m ) .
the data coverage is from 02/01/2001 until 24/03/2015 , for a total of 3711 data points .
all data were retrieved from datastream .
we analyze each time series using the methodology described in section [ sec : methodology ] .
the pdf was computed following the bp recipe , because it is the single method to introduce time - causality in pdf building .
although we have performed our analysis using embedding dimensions @xmath69 and 5 with time lag @xmath70 , we present the results only for @xmath71 , because it exhibits the best clarity for our explanations .
the other embedding dimensions results are similar .
we would like to know if the underlying generating process changes during the observation period . in order to evaluate such change
, we construct sliding windows .
the sliding windows work as follows : we consider the first @xmath72 data points and evaluate the two quantifiers , @xmath73 and @xmath74 .
then we move @xmath75 data points forward and compute the quantifiers of the new window with the following @xmath72 data points .
we continue this process until the end of the time series . after doing so , we obtain 170 estimation windows .
each window spans approximately one year , and the window step is approximately one month .
the initial and final dates of each estimation window is displayed in the electronic supplementary material esm - table 1 .
we also present in the electronic supplementary material the graphs corresponding to @xmath31-plane for the libor rates in gbp ( esm - fig . 1 )
, euro ( esm - fig .
2 ) , chf ( esm - fig .
3 ) , jpy ( esm - fig .
4 ) for the seven considered maturities : o / n , 1w , 2 m , 3 m , 6 m and 12 m , when they are evaluated on the 170 estimation windows respectively , given in this way a representation of the time evolution of them . from these figures , it can be observed that 1 m , 2 m and 3 m maturities have similar shape .
it can also noted that 6 m and 12 m are very similar to each other and their localization points are more concentrated .
we can summarize these results as : two distinct dynamics are detected .
on one side , 1 m , 2 m and 3 m maturities scatters throughout the plane . on the other side , o / n , 1w , 6 m and 12 m graphs are less disperse and their points are more concentrated on the low right corner . in the financial economics vocabulary , we can say that the latter maturities are more informationally efficient : their time series are closer to a random walk .
the middle maturities ( 1 m , 2 m , 3 m ) occupy places in the @xmath31plane that are consistent with fractional brownian motion or with chaotic dynamics .
this is the first evidence that we are dealing with a single interest rate but with distinct generating processes . in order to highlight the unequal behavior between the two groups of maturities , we count the number of points of the @xmath31plane that are closer to the low right corner and the number of points that are away from that corner .
we divide the points into two regions .
one , the most informational efficient region comprises the points where @xmath76 and @xmath77 .
the other is the complement of the @xmath31plane .
we display the results of the point counting in tables [ tab : pointsgbp ] to [ tab : pointsjpy ] .
these tables confirm the ocular inspection of the @xmath31plane ( see figures esm - fig . 1 to esm - fig .
whereas for 1 m , 2 m and 3 m maturities only between 33% to 39% of the points are in the efficient area , the percentages for the other maturities ranges 55% to 63% .
we recall that each point reflects the estimation of the quantifiers at a given moving window .
we would like to enquire if the aforementioned characteristics changed in an ordered way through time .
taking into account that we are studying the changes across time in the degree of informational efficiency of financial time series , we propose the following efficiency index , @xmath78 : @xmath79 = { \mathcal h}[p ] - { \mathcal f}[p]\ ] ] given that both quantifiers , @xmath73 and @xmath74 , are bounded between 0 and 1 , and that their behavior is expected opposite , the maximum value of our efficiency index is @xmath80 = 1 $ ] , when @xmath81=1 $ ] and @xmath82=0 $ ] and @xmath80 = -1 $ ] when @xmath83=0 $ ] and @xmath82=1 $ ] the most informational efficient behavior ( _ i.e. _ the most random behavior ) is when shannon entropy is maximized and fisher information minimized .
consequently , our efficiency index @xmath78 lies in the @xmath84 $ ] interval .
accordingly , our efficiency index can be represented in a color map as in fig .
[ fig : colormap ] ) . ]
we aim to assess the evolution of the efficiency @xmath78 through time . in order to do so
, we display the result for each currency in a color map where the @xmath85-axis is the temporal dimension ( _ i.e. _ the estimation windows ) and the @xmath86-axis are the different currencies .
the colors of the map reflects a degree of efficiency according to the color scale on the right of each graph and coincides with the color scale of fig .
[ fig : colormap ] .
the map corresponding gbp ( see fig . [
fig : colormapgbp ] ) clearly reflects two features .
first that o / n , 1w , 6 m and 12 m maturities has been behaving better that the middle maturities .
even more , 6 m and 12 m has been globally the most efficient during all the period . on contrary ,
1 m , 2 m , 3 m were informationally efficient until window 100 ( 05/08/2008 to 31/08/2009 ) and suffered a sudden shift in its stochastic behavior , that returns to its previous path around window 160 ( 12/03/2013 to 05/05/2014 ) .
additionally we can observe a yellow area around windows 6580 , that roughly correspond to years 2006 and 2007 .
the behavior detected by this color map is consistent with the alleged manipulation of libor rates reported in the press @xcite . according to our results
, libor rates suffered from some kind of inefficient behavior ( coherent with manipulation ) during the period 20062012 . in year 2013
the levels of informational efficiency recovered to levels similar to periods previous to the financial crisis . .
number of points in the inefficient ( @xmath87 ) and efficient ( @xmath88 ) regions for different maturities of libor gbp .
the information quantifiers were computing using bandt - pompe pdf with @xmath89 and @xmath34 .
[ cols= " > , > , > , > " , ] [ tab : pointsjpy ]
the first finding that we can extract from our data analysis is that the libor for all currencies and for maturities 1 m , 2 m and 3 m is governed by a different process than the other maturities .
the mentioned maturities usually reflect a lower level of informational efficiency than the other maturities .
the second important finding is that during the years 2007 - 2012 ( windows 120 - 160 ) there was a significant reduction in the informational efficiency of all markets , except the euro market .
this behavior seems to be contemporary to the uncovered manipulation announced in the newspapers @xcite .
we would like to highlight that our method is not intended to find manipulation , but rather to uncover changes in the hidden stochastic structure of data .
however , it is able to clearly discriminate areas of more deterministic behavior in the time series , which could be consistent with the alleged interest rate rigging .
snider ca , youle t , 2010 does the libor reflect banks borrowing costs ? .
available at _ social sicence research network _ ( ssrn ) : http://ssrn.com/abstract=1569603 or http://dx.doi.org/10.2139/ssrn.1569603 .
snchez - moreno p , yez rj , dehesa js , 2009 discrete densities and fisher information . in proceedings of the 14th international conference on difference equations and applications .
_ difference equations and applications .
_ istanbul , turkey : uur baheehir university publishing company , 291298 .
rosso oa , blanco s , jordanova j , kolev v , figliola a , schrmann m , baar e , 2001 wavelet entropy : a new tool for analysis of short duration brain electrical signals .
_ j. neurosci
. methods . _
* 105 * 6575 .
rosso oa , olivares f , zunino l , de micco l , aquino all , plastino a , larrondo ha , 2012 characterization of chaotic maps using the permutation bandt - pompe probability distribution .
j. b _ * 86 * 116129 .
soriano mc , zunino l , rosso , oa , fischer i , mirasso cr , 2011 time scales of a chaotic semiconductor laser with optical feedback under the lens of a permutation information analysis . _ ieee j. quantum electron .
_ * 47 * 252-261 . | this paper analyzes libor interest rates for seven different maturities and referred to operations in british pounds , euro , swiss francs and japanese yen , during the period years 2001 to 2015 .
the analysis is performed by means of two quantifiers derived from information theory : the permutation shannon entropy and the permutation fisher information measure .
an anomalous behavior in the libor is detected in all currencies except euro during the years 20062012 .
the stochastic switch is more severe in 1 , 2 and 3 months maturities .
given the special mechanism of libor setting , we conjecture that the behavior could have been produced by the manipulation that was uncovered by financial authorities .
we argue that our methodology is pertinent as a market overseeing instrument . *
pacs : * 89.65.gh econophysics ; 74.40.de noise and chaos |
recent neutrino oscillation experiments suggest the strong evidences of the tiny neutrino masses and the lepton flavor mixing @xcite-@xcite .
studies of the lepton flavor mixing matrix , which we call maki - nakagawa - sakata ( mns ) matrix @xcite , will give us important cues of the physics beyond the standard model .
one of the most important studies of the lepton - flavor mixing is the analysis of quantum corrections of the mns matrix @[email protected] , the majorana phase plays the crucial role for stabilities of the mixing angle against quantum corrections @xcite . in this paper
we analyze the energy - scale dependence of the mns matrix with the physical majorana phase in two generation neutrinos . according to the lsnd experiment@xcite
, the scenario of two heavy degenerate neutrinos can be realistic , where we can neglect the first generation effects in the energy - scale dependence of the mns matrix .
in this scenario , the mns matrix has two parameters , which are the mixing angle and the majorana phase .
the energy - scale dependence of them are completely determined by only one parameter @xmath0 which is the ratio between the diagonal elements of neutrino mass matrix .
the energy - scale dependence of @xmath0 can be easily solved .
when we use this parameter @xmath0 , we can easily obtain the analytic solution to the energy - scale dependence of physical parameters in the mns matrix which are the mixing angle and the majorana phase .
the energy - scale dependence of @xmath1 is also determined by only one parameter @xmath0 , where @xmath2 is the mass - squared difference between the second and the third generations and @xmath3 is the absolute value of the heaviest neutrino mass .
the energy - scale dependence of the majorana phase is clarified by using these analytic solutions .
the instability of the majorana phase causes in the same parameter region in which the mixing angle is unstable against quantum corrections .
the effective yukawa couplings in the lepton sector are given by @xmath5 where @xmath6 and @xmath7 are the su(2)@xmath8 doublet higgs bosons that give dirac masses to the up - type and down - type fermions , respectively .
@xmath9 is the @xmath10-th generation su(2)@xmath8 doublet lepton .
@xmath11 is the @xmath10-th generation right - handed charged - lepton .
the matrix @xmath4 induces the neutrino majorana mass matrix .
@xmath12 is the yukawa matrix of the charged - lepton . in this paper , we take the diagonal base of @xmath12 . once @xmath12 is taken diagonal at a certain scale , the diagonality of it is kept at all energies in the one - loop level .
this is because there are no lepton - flavor - mixing terms , except for @xmath4 in the minimal supersymmetric standard model ( mssm ) . in this base
the matrix @xmath4 is diagonalized as @xmath13 where @xmath14 is an unitary matrix ( the mns matrix ) and @xmath15 is the diagonalized mass matrix .
the mns matrix @xmath14 is parameterized as , @xmath16 where @xmath17 is the mixing angle and @xmath18 is the majorana phase .
the phases of @xmath19 and @xmath20 are unphysical parameters , which can be rotated out by the field redefinition .
since @xmath4 is complex and symmetric matrix , @xmath4 can be parameterized as @xcite @xmath21 in this parameterization , @xmath22 and @xmath23 are non - negative values and energy - scale independent parameters .
the energy - scale dependence of the @xmath4 can be controlled by two parameters @xmath0 and @xmath24 . since the @xmath24 is overall factor , the energy - scale dependence of the mixing angle and the majorana phase in the mns matrix
are governed by only one parameter @xmath0 .
the renormalization group equation for @xmath0 is given as @xcite @xmath25 where @xmath26 is the scaling parameter which is related to the renormalization scale @xmath27 as @xmath28 .
@xmath29 and @xmath30 are yukawa couplings of the charged - leptons , @xmath31 and @xmath27 , respectively .
we can obtain the solution to eq.([rger1 ] ) as @xmath32 where @xmath33 is a certain high energy scale .
when @xmath34 , where @xmath35 is the ratio of two vacuum expectation values of higgs bosons @xmath36 , the parameter @xmath37 can be obtained approximately @xcite as @xmath38 figure [ fig : epsilon ] shows the @xmath35 dependence of @xmath37 at @xmath39gev and @xmath40gev .
since the energy - scale dependence of @xmath0 is very small , the difference between the values of @xmath41 and @xmath42 is not so large @xcite .
for instance , in the case of @xmath43 and @xmath44 , @xmath37 is nearly equal to @xmath45 .
by using the energy - scale dependent parameters @xmath0 and @xmath24 , and the energy - scale independent parameters @xmath22 and @xmath23 , the masses of neutrinos , @xmath46 and @xmath47 , are given as @xmath48 the function @xmath49 is defined as @xmath50 where the function @xmath51 is defined as @xmath52 the mass - squared difference between neutrinos is given as @xmath53 the mass - squared difference scaled by the heaviest neutrino mass @xmath54 can be given as @xmath55 by using the parameters , @xmath22 , @xmath23 , and @xmath0 , the mixing angle @xmath17 and the majorana phase @xmath18 can be written as @xmath56 and @xmath57
the energy - scale dependent parameter is only @xmath0 in the physical parameters of eqs.([deg1 ] ) , ( [ mix1 ] ) and ( [ majo1 ] ) . thus studying the energy - scale dependence of the physical parameters is equivalent to studying the @xmath0 dependence of them . at the low energy - scale , the large
mixing between the second and the third generations , @xmath58 is favored by the atmospheric neutrino experiments @xcite .
this means that @xmath59 should be required at the low energy - scale from the eq.([mix1 ] ) .
there are following two cases which satisfy the eq.([eq : cond1 ] ) : \(a ) the parameter @xmath22 is very large such as @xmath60 in this case , the energy - scale dependence of the physical parameters is negligible , because the dominant component of the @xmath4 is the energy - scale independent parameter @xmath22 .
however , even if the parameter @xmath22 is enough large but finite , eq.([eq : cond2 ] ) is not satisfied in the vicinity of @xmath61 . in this parameter region ,
the large mixing is unstable against quantum corrections .
when @xmath0 is in the vicinity of the zero , the parameter @xmath22 must be infinitely large in order to satisfy eq.([eq : cond2 ] ) .
equation([eq : cond2 ] ) can be rewritten as @xmath62 where @xmath63 is the absolute value of the off - diagonal elements in the matrix @xmath4 .
when is satisfied , the large mixing is always preserved against quantum corrections @xcite .
this case corresponds with so - called pseudo - dirac type neutrino mass matrix @xcite .
\(b ) the parameter @xmath0 is nearly equal to 1 with @xmath64 .
we discuss the case of @xmath65 later . in this case , it is necessary that we study the @xmath0 dependence of the physical parameters in the vicinity of the @xmath66 , because mixing angle @xmath67 is not stable against quantum corrections in this region @xcite .
since the large @xmath22 makes the mixing angle to be stable against quantum corrections as we have seen in the case ( a ) , we study the region of @xmath68 . here
we input @xmath69 at @xmath70 scale .
figure [ fig : mixing1 ] shows the contour plot of @xmath71 at @xmath33 scale ( @xmath72 ) for the continuous changes of @xmath22 and @xmath23 .
the more the value deviates from 1 the larger the change of the mixing angle by quantum corrections becomes . as the values of @xmath23 approach @xmath73 and the parameter @xmath22 becomes small , the @xmath71 becomes unstable against quantum correction .
when we define @xmath74 and take the parameters , @xmath22 , @xmath37 and @xmath75 to be small , the left - hand side of eq.([eq : cond1 ] ) can be rewritten as @xmath76 this indicates that if the @xmath77 is much smaller than @xmath37 , eq.([eq : cond1 ] ) is not satisfied . therefor the mixing angle is unstable against the quantum corrections in this region .
this result is the same as that of ref .
@xcite . in order to understand the physical meaning of the parameters @xmath0(@xmath37 ) , @xmath22 and @xmath23(@xmath75 )
obviously , these parameters should be related to the other physical parameters , the majorana phase and the mass - squared difference .
especially , it is necessary to investigate the physical meaning of the parameter regions in which the mixing angle @xmath71 is unstable against quantum corrections .
figure [ fig : majorana1 ] shows the majorana phase @xmath18 in eq.([majo1 ] ) in the vicinity of @xmath78 for @xmath79 and @xmath80 cases , when we take @xmath81 and @xmath72 for both cases .
we set the @xmath82 at the @xmath70 scale .
figure [ fig : majorana1 ] shows that the majorana phase is not changed by the quantum corrections , when the @xmath23 is not close to @xmath73 . by using small parameter @xmath75 , the majorana phase is obtained as @xmath83 for @xmath66 . by using this equation
, we obtain the majorana phase at the @xmath70 scale for the large mixing . on the other hand , when the @xmath23 is close to the @xmath73 , the majorana phase is changed by the quantum corrections and it always becomes close to @xmath84 .
the mixing angle is unstable against quantum corrections in the region of small @xmath22 and @xmath85 .
comparing to figure [ fig : mixing1 ] and figure [ fig : majorana1 ] , we notice that the instability of the majorana phase causes in the same parameter region in which the mixing angle is unstable against quantum corrections .
figure [ fig : massdiff1 ] shows the contour plot of @xmath86 in eq.([deg1 ] ) at @xmath81 and @xmath72 for the continuous changes of @xmath22 and @xmath23 .
according to the figure [ fig : massdiff1 ] , the parameter region in which the mixing angle is unstable against quantum corrections is corresponding to the region in which the masses are more degenerate .
this result is consistent with that of ref .
@xcite .
finally , we discuss the special case of @xmath65 , where the @xmath0 dependence of the mixing angle and the majorana phase @xmath18 are quite different from the case of @xmath64 . in this case eq.([mix1 ] ) becomes @xmath87 this equation indicates that @xmath71 does not automatically take the maximal mixing at @xmath88 .
it is because the mixing angle becomes @xmath89 when @xmath66 .
thus the large mixing angle is only realized when @xmath90 ( case ( a ) ) where the large mixing is stable against quantum corrections .
for the majorana phase , we can obtain @xmath91 from eq.([majo1 ] ) at @xmath65 . in this case
, the majorana phase does not depend on @xmath0 and @xmath22 .
therefore the majorana phase is independent of the energy - scale .
in this paper we study the energy - scale dependence of the mns matrix with the physical majorana phase in two generation neutrinos . according to the lsnd experiment@xcite
, the scenario of two heavy degenerate neutrinos can be realistic and important , where the first generation effects are neglected in the energy - scale dependence of the mns matrix . in this case
, the mns matrix has two physical parameters , which are the mixing angle @xmath17 and the majorana phase @xmath18 .
the energy - scale dependence of these parameters are controlled by only one parameter @xmath0 which is the ratio of the diagonal elements in the neutrino mass matrix @xcite .
this @xmath0 also governs the mass - squared difference scaled by the heaviest neutrino mass .
we can easily solve the renormalization group equation of this parameter , @xmath0 .
then the analytic solutions to the energy - scale dependence of the physical parameters can be obtained .
especially , the energy - scale dependence of the majorana phase is clarified by using these analytic solutions .
the instability of the majorana phase causes in the same parameter region in which the mixing angle is unstable against quantum corrections .
one of author no is supported by the jsps research fellowship for young scientists , no.2996 .
homestake collaboration , b.t .
cleveland _ et al . _ , nucl .
+ ( proc . suppl . )
* b38 * , 47 ( 1995 ) ; + kamiokande collaboration , y. suzuki , nucl .
( proc . suppl . )
* b38 * , 54 ( 1995 ) ; + gallex collaboration , p. anselmann _ et al .
_ , ; + sage collaboration , j.n .
abdurashitov _ et al .
_ , ; + super - kamiokande collaboration , ; ; hep - ex/9903034 . p.h .
chankowski and z. pluciennik , ; + k.s .
babu , c.n .
leung and j. pantaleone , ; + p.h .
chankowski , w. krolikowski and s. pokorski , hep - ph/9910231 ; + j.a .
casas , j.r .
espinosa , a. ibarra and i. navarro , hep - ph/9910420 .
m. tanimoto , ; + j. ellis , g.k .
leontaris , s. lola and d.v .
nanopoulos , ; + s. lola , hep - ph/9903203 ; + j.a .
casas , j.r .
espinosa , a. ibarra and i. navarro , hep - ph/9904395 , hep - ph/9905381 , hep - ph/9906281 ; + m. carena , j. ellis , s. lola and c.e.m .
wagner , hep - ph/9906362 . | in the case of two generation neutrinos , the energy - scale dependence of the lepton - flavor mixing matrix with majorana phase can be governed by only one parameter @xmath0 , which is the ratio between the diagonal elements of neutrino mass matrix . by using this parameter @xmath0
, we derive the analytic solutions to the renormalization group equations of the physical parameters , which are the mixing angle , majorana phase , and the ratio of the mass - squared difference to the mass squared of the heaviest neutrino .
the energy - scale dependence of the majorana phase is clarified by using these analytic solutions .
the instability of the majorana phase causes in the same parameter region in which the mixing angle is unstable against quantum corrections .
hep - ph/9911481 + dpnu-99 - 31 + kek - th-662 |
compactness is a fundamental and important property in both theory and applications @xcite .
compactness also plays an important role in the applications of fuzzy sets @xcite .
the compact criteria attracts much attention .
it s well - known that arzel
ascoli theorem(s ) provide compact criteria in classic analysis and topology .
there also exist many important and interesting works including @xcite which characterized compactness in fuzzy sets spaces equipped with different topologies . since diamond and kloeden @xcite introduced @xmath4 metric which is a @xmath0-type metric , it has become one of the most often used convergence structure on fuzzy sets .
naturally , people have started to consider the characterizations of compactness in fuzzy sets spaces with @xmath4 metric .
diamond and kloeden @xcite gave compact criteria of fuzzy number space @xmath5 with @xmath4 metric .
ma @xcite modified the characterization given by diamond and kloeden . wu and
zhao @xcite pointed out that characterization proposed by ma is still wrong , and they gave a right characterization in @xmath6 .
convexity is a very useful property .
star - shapedness is a natural extension of convexity . of cause ,
research on fuzzy counterparts of star - shaped sets has aroused the interest of people @xcite .
diamond @xcite introduced the set of fuzzy star - shaped numbers which is denoted by @xmath7 .
he characterized the compact sets in @xmath8 , where @xmath9 denotes the set of all the fuzzy star - shaped numbers with respect to the origin . @xmath5 and @xmath9 do not include each other .
they both are subsets of @xmath7 .
wu and zhao @xcite pointed out that this characterization is wrong and gave a characterization of compactness in @xmath8 . based on the results in @xcite , zhao and
wu @xcite further proposed a characterization of compactness in @xmath10 . in these discussions
, it is found that the concept `` @xmath11-mean equi - left - continuous '' proposed by diamond and kloeden @xcite plays an important role in establishing and illustrating characterizations of compactness in fuzzy sets spaces with @xmath4 metric .
compare the characterizations in @xcite to arzel
ascoli theorem , we find that the latter provides the compact criteria by characterizing the totally bounded sets while the former does not seem to characterize the totally bounded sets . since , in metric space , totally boundedness is a key feature of compactness , it is a natural and important problem to consider how to characterize totally bounded sets in spaces of fuzzy sets with @xmath4 metric ?
qiu et al.@xcite introduced @xmath12 , the set of all general fuzzy star - shaped numbers , which take fuzzy star - shaped numbers as a special case .
it is a more natural fuzzy extension of star - shaped sets in some sense .
so this has caused an important problem : how to characterize compact sets in @xmath13 ? as diamond @xcite pointed out , the four spaces @xmath6 , @xmath8 , @xmath10 and @xmath13 are not complete .
krtschmer @xcite presented the completion of @xmath6 which is described by the support functions of fuzzy numbers .
it raises a natural and basic problem : what are the completions of all the rest spaces ? in this paper , we want to answer the above three questions , they are relevant to each other .
an interesting fact is that our discussion can be put in a more general framework which does not have any assumptions of convexity or star - shapedness .
so we consider the set of all normal , upper semi - continuous , compact - support fuzzy sets on @xmath3 , which is denoted by @xmath14 .
further we introduce @xmath15 , the @xmath0-type extension of @xmath14 .
all the fuzzy sets spaces mentioned in this paper are subspaces of @xmath16 .
the remainder part of this paper is organized as follows . since @xmath4 metric are based on the well - known hausdorff metric , section 2 introduces and discusses some properties of hausdorff metric . in section 3 , we recall and introduce some concepts and results of fuzzy sets related to our paper . then , in section 4 , we present characterizations of relatively compact sets , totally bounded sets and compact sets in @xmath17 .
section 5 shows that @xmath15 is in fact the completion of @xmath18 according to @xmath4 metric .
then it gives the characterizations of relatively compact sets , totally bounded sets and compact sets in @xmath19 .
these are key results of this paper . based on the conclusions in sections 4 and 5 and discussions on convexity and star - shapedness of fuzzy sets , in section 6
, it shows that @xmath20 , @xmath21 , @xmath22 and @xmath23 are just the completions of @xmath24 , @xmath6 , @xmath10 and @xmath13 , respectively .
we clarify the relation among the ten fuzzy sets spaces discussed in this paper . as consequences of preceding results
, it obtains characterizations of totally bounded sets , relatively compact sets and compact sets in these spaces .
let @xmath25 be the set of all natural numbers , @xmath26 be the set of all rational numbers , @xmath27 be @xmath28-dimensional euclidean space , @xmath29 be the set of all the nonempty compact and convex sets in @xmath27 , @xmath30 be the set of all nonempty compact set in @xmath27 , and @xmath31 be the set of all nonempty closed set in @xmath27 .
the well - known hausdorff metric @xmath32 on @xmath33 is defined by : @xmath34 for arbitrary @xmath35 , where @xmath36 @xcite [ kcs ] @xmath37 is a complete metric space in which @xmath38 and @xmath39 are closed subsets .
hence , @xmath38 and @xmath29 are also complete metric spaces .
@xcite [ sca ] a nonempty subset @xmath40 of @xmath41 is compact if and only if it is closed and bounded in @xmath41 .
@xcite [ mce ] let @xmath42 satisfy @xmath43 then @xmath44 and @xmath45 . on the other hand , if @xmath46 and @xmath47 , then @xmath45 .
a set @xmath48 is said to be star - shaped relative to a point @xmath49 if for each @xmath50 , the line @xmath51 joining @xmath52 to @xmath53 is contained in @xmath54 .
the kernel ker@xmath54 of @xmath54 is the set of all points @xmath49 such that @xmath55 for each @xmath50 .
the symbol @xmath56 is used to denote all the star - shaped sets in @xmath27 .
obviously , @xmath57 .
it can be checked that @xmath58 for all @xmath59 .
we say that a sequence of sets @xmath60 converges to @xmath61 , in the sense of kuratowski , if @xmath62 where @xmath63 in this case , we ll write simply @xmath64 .
the following two known propositions discuss the relation of the convergence induced by hausdorff metric and the convergence in the sense of kuratowski .
the readers can see @xcite for details .
[ hms ] suppose that @xmath65 , @xmath66 , @xmath67 , are nonempty compact sets in @xmath27 .
then @xmath68 as @xmath69 implies that @xmath70 . [ klhe ]
suppose that @xmath65 , @xmath66 , @xmath67 , are nonempty compact sets in @xmath27 and that @xmath66 , @xmath67 , are connected sets . if @xmath71 , then @xmath68 as @xmath69 . [ ksc ] @xmath56 is a closed set in @xmath72 .
* proof * suppose that @xmath73 , @xmath74 and @xmath68 as @xmath69 .
in the following , we will prove that @xmath75 .
choose @xmath76 , @xmath67 , then there exists an @xmath77 such that @xmath78 for all @xmath79 , where @xmath80 .
note that @xmath40 is a compact set , we know that there is a subsequence @xmath81 of @xmath82 such that @xmath83 .
so @xmath84 , it then follows from proposition [ hms ] that @xmath85 .
now , we show that @xmath65 is star - shaped and @xmath86 .
it suffices to show that @xmath87 for all @xmath88 and @xmath89 $ ] .
in fact , given @xmath88 , since @xmath90 , there is a sequence @xmath91 such that @xmath92 .
hence , for each @xmath89 $ ] , @xmath93 and thus , by proposition [ hms ] , @xmath94 .
@xmath95 [ kef ] let @xmath65 , @xmath66 be star - shaped sets , @xmath67 . if @xmath68 , then @xmath96 .
* proof * from the proof of theorem [ ksc ] , we get the desired results .
we do not know whether theorem [ ksc ] and corollary [ kef ] are known conclusions , so we give our proofs here .
in this section , we recall and introduce various spaces of fuzzy sets including fuzzy numbers space , fuzzy star - shaped numbers space and general fuzzy star - shaped numbers space .
it will be shown in this paper that these spaces form pairs of original spaces and their completions with @xmath0-metric .
some basic properties of these spaces are discussed .
we use @xmath97 to represent all fuzzy subsets on @xmath27 , i.e. functions from @xmath27 to @xmath98 $ ] . for details
, we refer the readers to references @xcite .
@xmath99 can be embedded in @xmath100 , as any @xmath101 can be seen as its characterization function , i.e. the fuzzy set @xmath102 for @xmath103 , let @xmath104_{{\alpha}}$ ] denote the @xmath105-cut of @xmath65 , i.e. @xmath106_{{\alpha}}=\begin{cases } \{x\in \mathbb{r}^m :
u(x)\geq { \alpha}\ } , & \ { \alpha}\in(0,1 ] , \\ { \rm supp}\ , u=\overline{\{x \in \mathbb{r}^m : u(x)>0\ } } , & \ { \alpha}=0 . \end{cases}\ ] ] for @xmath103 , we suppose that + ( 1 ) @xmath65 is normal : there exists at least one @xmath107 with @xmath108 ; + ( 2 ) @xmath65 is upper semi - continuous ; + ( 3 - 1 ) @xmath65 is fuzzy convex : @xmath109 for @xmath110 and @xmath111;$ ] + ( 3 - 2 ) @xmath65 is fuzzy star - shaped , i.e. , there exists @xmath112 such that @xmath65 is fuzzy star - shaped with respect to @xmath52 , namely , @xmath113 for all @xmath114 ; + ( 3 - 3 ) given @xmath115 $ ] , then there exists @xmath116_\lambda$ ] such that @xmath117_\lambda$ ] for all @xmath118_\lambda$ ] ; + ( 4 - 1 ) @xmath104_0 $ ] is a bounded set in @xmath27 ; + ( 4 - 2 ) @xmath119_{\alpha } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } < + \infty $ ] , where @xmath120 and @xmath121 denotes the origin of @xmath27 ; + ( 4 - 3 ) @xmath104_{\alpha}$ ] is a bounded set in @xmath27 when @xmath122 . *
if @xmath65 satisfies ( 1 ) , ( 2 ) , ( 3 - 1 ) and ( 4 - 1 ) , then @xmath65 is a fuzzy number .
the set of all fuzzy numbers is denoted by @xmath5 . *
if @xmath65 satisfies ( 1 ) , ( 2 ) , ( 3 - 2 ) and ( 4 - 1 ) , then we say @xmath65 is a fuzzy star - shaped number .
the set of all fuzzy star - shaped numbers is denoted by @xmath7 . *
if @xmath65 satisfies ( 1 ) , ( 2 ) , ( 3 - 3 ) and ( 4 - 1 ) , then we say @xmath65 is a general fuzzy star - shaped number .
the set of all general fuzzy star - shaped numbers is denoted by @xmath12 .
@xmath123 can be embedded in @xmath5 , as any @xmath124 can be viewed as the fuzzy number @xmath125 we can see that if @xmath126 , then @xmath127}\mbox{ker}\;[u]_\lambda \not= \emptyset$ ] , however this inequality may not hold when @xmath128 .
so @xmath129 .
the definition of fuzzy star - shaped numbers was given by wu and zhao @xcite .
the concept of general fuzzy star - shaped numbers was given by qiu et al .
@xcite .
diamond and kloeden @xcite introduced the @xmath4 distance ( @xmath130 ) on @xmath12 which are defined by @xmath131_{\alpha } , [ v]_{\alpha } ) ^p \ ; d\alpha \right)^{1/p}\ ] ] for all @xmath132 .
therein , they pointed out that @xmath4 is a metric on @xmath12 and that the three spaces @xmath6 , @xmath10 and @xmath133 are not complete .
krtschmer @xcite has given the completion of @xmath6 which is described by the support functions of fuzzy numbers . to discuss properties of the above three spaces more clearly
, we need to consider @xmath0type noncompact fuzzy sets . by relaxing assumption ( 4 - 1 ) a little to get ( 4 - 2 )
, we obtain the following @xmath0type noncompact fuzzy sets .
suppose that @xmath134 .
* if @xmath65 satisfies ( 1 ) , ( 2 ) , ( 3 - 1 ) and ( 4 - 2 ) , then we say @xmath65 is a @xmath0-type non - compact fuzzy number .
the set of all such fuzzy numbers is denoted by @xmath135 . * if @xmath65 satisfies ( 1 ) , ( 2 ) , ( 3 - 2 ) and ( 4 - 2 ) , then we say @xmath65 is a @xmath0-type non - compact fuzzy star - shaped number .
the set of all such fuzzy star - shaped numbers is denoted by @xmath136 . * if @xmath65 satisfies ( 1 ) , ( 2 ) , ( 3 - 3 ) and ( 4 - 2 ) , then we say @xmath65 is a @xmath0-type non - compact general fuzzy star - shaped number .
the set of all such general fuzzy star - shaped numbers is denoted by @xmath137 . clearly , @xmath138 , @xmath139 , @xmath140 and @xmath141 .
the discussions in this paper can be put in a more general framework without any assumptions of convexity or star - shapedness .
so we consider the following types of fuzzy sets .
suppose that @xmath134 .
* if @xmath65 satisfies ( 1 ) , ( 2 ) and ( 4 - 1 ) , then @xmath65 is a normal upper semi - continuous compact - support fuzzy set on @xmath3 .
the set of all such fuzzy sets is denoted by @xmath14 . *
if @xmath65 satisfies ( 1 ) , ( 2 ) and ( 4 - 2 ) , then @xmath65 is a normal upper semi - continuous @xmath0-type non - compact - support fuzzy set on @xmath3 .
the set of all such fuzzy sets is denoted by @xmath15 . *
if @xmath65 satisfies ( 1 ) , ( 2 ) and ( 4 - 3 ) , then @xmath65 is a normal upper semi - continuous non - compact - support fuzzy set on @xmath3 .
the set of all such fuzzy sets is denoted by @xmath142 .
note that @xmath143 satisfies assumption ( 2 ) is equivalent to @xmath104_{\alpha}\in c ( \mathbb{r}^m)$ ] for all @xmath144 $ ] .
so @xmath145 .
it is easy to see that @xmath146 and @xmath147 are subsets of @xmath14 and @xmath148 , respectively .
it can be checked that the @xmath4 distance , @xmath120 , ( see eq . )
is also a metric on @xmath15 . but @xmath4 distance is not a metric on @xmath149 because @xmath150 may equal @xmath151 for some @xmath152 .
all the fuzzy sets spaces mentioned in this paper are subspaces of @xmath16 . in this paper , we show that @xmath153 is the completion of @xmath154 .
then we find that @xmath155 , @xmath156 and @xmath23 are just the completion of @xmath6 , @xmath157 and @xmath158 respectively .
the following lemma shows that for each @xmath159 , the only possible unbounded cut - set is the 0-cut , @xmath104_0 $ ] .
[ ulce ] given @xmath159 , then for each @xmath160 $ ] , @xmath104_{\alpha}$ ] is a compact set .
* proof * since @xmath104_{\alpha}$ ] is a closed set , it only need to show that @xmath104_{\alpha}$ ] is a bounded set , i.e. @xmath161_{\alpha } , \{0\ } ) < + \infty$ ] .
we proceed by contradiction .
if @xmath161_{\alpha } , \{0\})=+\infty$ ] . note that @xmath161_\beta , \{0\ } ) \geq h([u]_{\alpha } , \{0\ } ) $ ] when @xmath162 .
then @xmath163_{\alpha } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } \geq \left ( \int_0^\alpha h([u]_{\alpha } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } = + \infty , $ ] which contradicts the fact that @xmath159 .
@xmath95 [ act ] by lemma [ ulce ] , we know that @xmath104_{\alpha}\in k_c ( \mathbb{r}^m)$ ] for all @xmath164 and @xmath160 $ ] , and that @xmath104_{\alpha}\in k_s ( \mathbb{r}^m)$ ] for all @xmath165 and @xmath160 $ ] .
denote @xmath166 } \mbox{ker}\ ; [ u]_{\alpha}$ ] for @xmath165 ( also see @xcite ) .
it is easy to check that , given @xmath165 , then @xmath167 if and only if @xmath168 .
the following representation theorem is used widely in the theory of fuzzy numbers .
@xcite [ nr]given @xmath169 then + ( 1 ) @xmath104_{\lambda}\in k_c(\mathbb{r}^m)$ ] for all @xmath111 $ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] .
moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath173 then there exists a unique @xmath174 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111.$ ] similarly , we can obtain representation theorems for @xmath7 , @xmath12 , @xmath18 , @xmath135 , @xmath136 , @xmath137 and @xmath175 which will be used in the sequel .
[ rslp ] given @xmath176 then + ( 1 ) @xmath104_{\lambda}\in k_s(\mathbb{r}^m)$ ] for all @xmath111 $ ] , and @xmath177 } \mbox{ker}\ ; [ u]_\alpha \not= \emptyset$ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] .
moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath173 , then there exists a unique @xmath178 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111 $ ] .
[ rgslp ] given @xmath179 then + ( 1 ) @xmath104_{\lambda}\in k_s(\mathbb{r}^m)$ ] for all @xmath111 $ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] .
moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath173 , then there exists a unique @xmath180 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111 $ ] .
[ rfbe ] given @xmath181 then + ( 1 ) @xmath104_{\lambda}\in k(\mathbb{r}^m)$ ] for all @xmath111 $ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] .
moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath173 , then there exists a unique @xmath182 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111 $ ] .
[ rep ] given @xmath183 then + ( 1 ) @xmath104_{\lambda}\in k_c(\mathbb{r}^m)$ ] for all @xmath170 $ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] ; + ( 4 ) @xmath119_{\alpha } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } < + \infty $ ]
. moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath184 , then there exists a unique @xmath164 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111 $ ] .
[ rs0p ] given @xmath185 then + ( 1 ) @xmath104_{\lambda}\in k_s(\mathbb{r}^m)$ ] for all @xmath170 $ ] and @xmath177 } \mbox{ker}\ ; [ u]_\alpha \not= \emptyset$ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] ; + ( 4 ) @xmath119_{\alpha } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } < + \infty $ ] .
moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath184 , then there exists a unique @xmath167 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111 $ ] .
[ rsp ] given @xmath186 then + ( 1 ) @xmath104_{\lambda}\in k_s(\mathbb{r}^m)$ ] for all @xmath170 $ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] ; + ( 4 ) @xmath119_{\alpha } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } < + \infty $ ] .
moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath184 , then there exists a unique @xmath187 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111 $ ] .
[ rfbp ] given @xmath188 then + ( 1 ) @xmath104_{\lambda}\in k(\mathbb{r}^m)$ ] for all @xmath170 $ ] ; + ( 2 ) @xmath104_{\lambda}=\bigcap_{\gamma<\lambda}[u]_\gamma$ ] for all @xmath170 $ ] ; + ( 3 ) @xmath104_0=\overline{\bigcup_{\gamma>0}[u]_\gamma}$ ] ; + ( 4 ) @xmath119_{\alpha } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } < + \infty $ ] .
moreover , if the family of sets @xmath171\}$ ] satisfy conditions @xmath172 through @xmath184 , then there exists a unique @xmath189 such that @xmath104_{{\lambda}}=v_\lambda$ ] for each @xmath111 $ ] .
in this section , we present a characterization of relatively compact sets in fuzzy sets space @xmath190 . based on this , we then give characterizations of totally bounded sets and compact sets in @xmath190 .
the following concepts are important to establish and illustrate the characterizations .
@xcite a set @xmath191 is said to be uniformly @xmath11-mean bounded if there is a constant @xmath192 such that @xmath193 for all @xmath194 .
we can see that @xmath40 is uniformly @xmath11-mean bounded is equivalent to @xmath40 is a bounded set in @xmath195 .
@xcite let @xmath196 .
if for given @xmath197 , there is a @xmath198 such that for all @xmath199 @xmath200_\alpha , [ u]_{\alpha - h})^p \;d\alpha \right)^{1/p } < \varepsilon , \ ] ] where @xmath201 , then we say @xmath65 is @xmath11-mean left - continuous .
suppose that @xmath40 is a nonempty set in @xmath175 .
if the above inequality holds uniformly for all @xmath194 , then we say @xmath40 is @xmath11-mean equi - left - continuous .
ma @xcite use @xmath202 to denote the fuzzy set @xmath202 induced by @xmath103 which is defined as follows : @xmath203 diamond @xcite characterized the compact sets in @xmath6 and @xmath8 , where @xmath9 denotes the set of all fuzzy star - shaped number @xmath65 with respect to @xmath121 , i.e. @xmath204 .
obviously , @xmath205 and @xmath206 .
@xcite [ dempc ] a closed set @xmath40 of @xmath6 is compact if and only if : + ( 1 ) @xmath207_0 : u\in u\}$ ] is bounded in @xmath208 ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous .
@xcite [ dsm0pc ] a closed set @xmath40 of @xmath8 is compact if and only if : + ( 1 ) @xmath207_0 : u\in u\}$ ] is bounded in @xmath208 ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous .
ma @xcite modified proposition [ dempc ] as follows .
@xcite [ mgep ] a closed set @xmath40 of @xmath6 is compact if and only if : + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) for @xmath209 , if @xmath210 converges to @xmath211 in @xmath4 metric for any @xmath212 , then there exists a @xmath213 such that @xmath214 .
wu and zhao @xcite pointed out the above three characterizations of compactness are wrong , and gave compactness criteria in spaces @xmath6 and @xmath8 .
@xcite [ gep ] a closed set @xmath40 of @xmath6 is compact if and only if : + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) let @xmath215 be a decreasing sequence in @xmath216 $ ] converging to zero .
for @xmath209 , if @xmath217 converges to @xmath218 in @xmath4 metric , then there exists a @xmath213 such that @xmath219_{\alpha}=[u(r_i)]_{\alpha}$ ] when @xmath220 .
@xcite [ gs0p ] a closed set @xmath40 of @xmath8 is compact if and only if : + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) let @xmath215 be a decreasing sequence in @xmath216 $ ] converging to zero . for @xmath209 ,
if @xmath217 converges to @xmath221 in @xmath4 metric , then there exists a @xmath222 such that @xmath219_{\alpha}=[u(r_i)]_{\alpha}$ ] when @xmath220 .
based on proposition [ gs0p ] , zhao and wu @xcite further presented compactness criteria of @xmath10 .
@xcite[gsp ] a closed set @xmath40 of @xmath10 is compact if and only if : + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) let @xmath215 be a decreasing sequence in @xmath216 $ ] converging to zero . for @xmath209 , if @xmath217 converges to @xmath223 in @xmath4 metric , then there exists a @xmath224 such that @xmath219_{\alpha}=[u(r_i)]_{\alpha}$ ] when @xmath220 .
compare propositions [ gep ] , [ gs0p ] , [ gsp ] with arzel
ascoli theorem , we notice that the latter provides the compact criteria by characterizing the totally bounded set while the former does not seem to do so .
since totally boundedness is a key feature of compactness , it is natural and important to consider characterizations of totally bounded sets . in this section ,
we consider characterizations of totally boundedness , relatively compactness and compactness in @xmath190 .
some fundamental conclusions and concepts in classic analysis and topology are listed below , which are useful in this paper .
the readers can see @xcite for details . * * lebesgue s dominated convergence theorem*. let @xmath225 be a sequence of integrable functions that converges almost everywhere to a function @xmath226 , and suppose that @xmath225 is dominated by an integrable function @xmath227 .
then @xmath226 is integrable , and @xmath228 . * * fatou s lemma*. let @xmath225 be a sequence of nonnegative integrable functions that converges almost everywhere to a function @xmath226 , and if the sequence @xmath229 is bounded above , then @xmath226 is integrable and @xmath230 * * absolute continuity of lebesgue integral*. suppose that @xmath226 is lebesgue integrable on @xmath231 , then for arbitrary @xmath197 , there is a @xmath232 such that @xmath233 whenever @xmath234 and @xmath235 . *
* minkowski s inequality*. let @xmath236 , and let @xmath237 be measurable functions on @xmath238 such that @xmath239 and @xmath240 are integrable .
then @xmath241 is integrable , and minkowski s inequality @xmath242 holds . *
a relatively compact subset @xmath243 of a topological space x is a subset whose closure is compact .
* let @xmath244 be a metric space .
a set @xmath40 in @xmath245 is totally bounded if and only if for each @xmath197 , it contains a finite @xmath246-approximation , where an @xmath246-approximation to @xmath40 is a subset @xmath247 of @xmath40 such that @xmath248 for each @xmath249 .
* let @xmath244 be a metric space .
then a set @xmath40 in @xmath245 is relatively compact implies that it is totally bounded . for subsets of a complete metric space
these two meanings coincide .
thus @xmath244 is a compact space iff @xmath245 is totally bounded and complete .
firstly , we need some lemmas .
[ unes ] suppose that @xmath40 is a bounded set in @xmath190 , then @xmath207_{\alpha } : u\in u\}$ ] is a bounded set in @xmath208 for each @xmath250 . * proof * if there exists an @xmath251 such that @xmath207_{{\alpha}_0 } : u\in u\}$ ] is not a bounded set in @xmath208 .
then there is a @xmath194 such that @xmath104_{{\alpha}_0 } \notin k(\mathbb{r}^m)$ ] or @xmath207_{{\alpha}_0 } : u\in u\}$ ] is a unbounded set in @xmath252 . for both cases ,
there exist @xmath253 such that @xmath254_{{\alpha}_0 } , \{0\ } ) > n\cdot ( \frac{1}{{\alpha}_0})^{1/p}$ ] when @xmath67 , and hence @xmath255_{{\alpha } } , \{0\ } ) ^p \ ; d\alpha \right)^{1/p } > n,$ ] which contradicts the boundness of @xmath40 .
@xmath95 lemma [ ulce ] can be seen as a corollary of lemma [ unes ] .
[ lcf ] suppose that @xmath196 and @xmath144 $ ] , then @xmath161_{\alpha } , [ u]_{\beta } ) \to 0 $ ] as @xmath256 .
* proof * the desired result follows immediately from theorem [ rfbp ] and proposition [ mce ] . @xmath95 [ uplc ] if @xmath196 , then @xmath65 is @xmath257mean left - continuous . * proof * given @xmath197 .
note that @xmath196 , from the absolute continuity of lebesgue integral , we know there exists an @xmath258 such that @xmath259_\alpha , \{0\})^p \;d\alpha \right)^{1/p } \leq \varepsilon/3.\ ] ] by lemma [ ulce ] , @xmath104_{\frac{h_1}{2 } } \in k(\mathbb{r}^m)$ ] , then there exists an @xmath192 such that @xmath161_{\frac{h_1}{2 } } , \{0\})<m$ ] .
this yields that @xmath260_\alpha , [ u]_{\alpha - h } ) \leq h([u]_\alpha , \{0\ } ) + h ( [ u]_{\alpha - h } , \{0\ } ) \leq 2m\ ] ] for all @xmath261 and @xmath262 . by lemma
[ lcf],we know that @xmath161_\alpha , [ u]_{\alpha - h } ) \rightarrow 0 $ ] when @xmath263 , it then follows from the lebesgue s dominated convergence theorem and that @xmath264_\alpha , [ u]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } \rightarrow 0\ ] ] when @xmath265 .
thus there exists an @xmath266 such that @xmath267_\alpha , [ u]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } < \varepsilon/3\ ] ] for all @xmath268 .
now combined and , we know that , for all @xmath269 , @xmath270_\alpha , [ u]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } \nonumber \\ & \leq \left(\int_h^{h_1 } h([u]_\alpha , [ u]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } + \left(\int_{h_1}^{1 } h([u]_\alpha , [ u]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } \nonumber \\ & \leq \left(\int_h^{h_1 } h([u]_\alpha , \{0\ } ) ^p \;d\alpha \right)^{1/p } + \left(\int_h^{h_1 } h ( [ u]_{\alpha - h } , \{0\ } ) ^p \;d\alpha \right)^{1/p } + \varepsilon/3 \nonumber \\ & \leq \left(\int_0^{h_1 } h([u]_\alpha , \{0\ } ) ^p \;d\alpha \right)^{1/p } + \left(\int_0^{h_1 } h ( [ u]_{\alpha } , \{0\ } ) ^p \;d\alpha \right)^{1/p } + \varepsilon/3 \nonumber \\ & \leq \varepsilon/3 + \varepsilon/3 + \varepsilon/3 = \varepsilon.\end{aligned}\ ] ] from the arbitrariness of @xmath246 , we know that @xmath65 is @xmath257mean left - continuous .
@xmath95 now , we arrive at the main results of this section .
[ pcn ] @xmath40 is a relatively compact set in @xmath271 if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous . *
proof * _ * necessity . * _ if @xmath40 is a relatively compact set in @xmath271 . since @xmath271 is a metric space , it follows immediately that @xmath40 is a bounded set in @xmath271 , i.e. @xmath40 is uniformly @xmath11-mean bounded .
now we prove that @xmath40 is @xmath257mean equi - left - continuous .
given @xmath197 .
since @xmath40 is a relatively compact set , there exists an @xmath272-net @xmath273 of @xmath40 . from lemma [ uplc ] , we know that @xmath274 is @xmath11-mean equi - left - continuous
. hence there exists @xmath232 such that @xmath275_\alpha , [ u_k]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } \leq \varepsilon/3.\ ] ] for all @xmath276 and @xmath277 .
given @xmath194 , there is an @xmath278 such that @xmath279 , and thus , by , we know that for all @xmath276 , @xmath280_\alpha , [ u]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } \nonumber \\ & \leq \left(\int_h^1 h([u]_\alpha , [ u_k]_{\alpha } ) ^p \;d\alpha \right)^{1/p } + \left(\int_h^1 h([u_k]_\alpha , [ u_k]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } + \left(\int_h^1 h([u_k]_{\alpha - h } , [ u]_{\alpha - h } ) ^p \;d\alpha \right)^{1/p } \nonumber \\ & \leq \varepsilon/3 + \varepsilon/3 + \varepsilon/3 = \varepsilon.\end{aligned}\ ] ] from the arbitrariness of @xmath246 and @xmath194 , we obtain that @xmath40 is @xmath11-mean equi - left - continuous .
_ * sufficiency . * _ if @xmath40 satisfies ( 1 ) and ( 2 ) . to show @xmath40
is a relatively compact set , it suffices to find a convergent subsequence of an arbitrarily given sequence in @xmath40 .
let @xmath281 be a sequence in @xmath40 . to find a subsequence @xmath282 of @xmath281 which converges to @xmath283 according to @xmath4 metric
, we split the proof into three steps . * step 1 .
* find a subsequence @xmath282 of @xmath281 and @xmath284 such that @xmath285_\alpha , [ v]_\alpha ) { \stackrel } { \mbox{a.e . } } { \to } 0 \ ( [ 0,1 ] ) .\ ] ] since @xmath40 is uniformly @xmath257mean bounded , by lemma [ unes ] , @xmath207_\alpha : u\in u\}$ ] is a bounded set in @xmath38 for each @xmath144 $ ] .
thus , by proposition [ sca ] , for each @xmath250 , @xmath207_\alpha : u\in u\}$ ] is a relatively compact set in @xmath208 .
arrange all rational numbers in @xmath216 $ ] into a sequence @xmath286 .
then @xmath281 has a subsequence @xmath287 such that @xmath288_{q_1}\}$ ] converges to @xmath289 , i.e. @xmath290_{q_1 } , u_{q_1 } ) \to 0 $ ] .
if @xmath291 have been chosen , we can choose a subsequence @xmath292 of @xmath293 such that @xmath294 _
{ q_{k+1 } } \}$ ] converges to @xmath295 .
thus we obtain nonempty compact sets @xmath296 . with @xmath297
whenever @xmath298 .
put @xmath299 for @xmath67 .
then @xmath282 is a subsequence of @xmath281 and @xmath300_{q_k } , u_{q_k } ) \to 0 \ \ \ \mbox{as } \ \ \ n\to \infty\ ] ] for @xmath301 .
define @xmath302 \ } $ ] as follows : @xmath303\hbox { ; } \\
\overline { \bigcup_{\alpha \in ( 0,1 ] } v_\alpha } , & \
\alpha=0 \hbox{. } \end{array } \right.\ ] ] then @xmath304 , @xmath305 $ ] , have the following properties : 1 .
@xmath306 for all @xmath170 $ ] ; 2 .
@xmath307 for all @xmath170 $ ] ; 3 .
@xmath308 .
in fact , by proposition [ mce ] , we obtain that @xmath309 for all @xmath310 $ ] .
thus property ( 1 ) is proved .
properties ( 2 ) and ( 3 ) follow immediately from the definition of @xmath304 .
define a function @xmath311 $ ] by @xmath312 then @xmath313 is a fuzzy set on @xmath27 . from properties ( 1 ) , ( 2 ) and ( 3 ) of @xmath304 , we know that @xmath314_{\alpha}=v_{\alpha}.\ ] ] so @xmath315
. clearly if the following statements ( ) and ( ) are true , then we obtain , i.e. @xmath316_\alpha , [ v]_\alpha ) { \stackrel } { \mbox{a.e . } } { \to } 0 \ ( [ 0,1 ] ) $ ] . 1
. @xmath317 is at most countable , where @xmath318_{\alpha}\}$ ] , where @xmath319_{\beta } } $ ] .
2 . if @xmath320 , then @xmath321_\alpha , [ v]_\alpha ) \to 0 \ \hbox{as } \ n\to \infty.\ ] ] firstly , we show assertion ( ) .
let @xmath322_\alpha \nsubseteq \overline { \{v > \alpha\ } } \ ; \}$ ] .
notice that @xmath323 ( in fact , it can be checked that @xmath324 ) . by the conclusion in appendix of @xcite , @xmath325 is at most countable .
so @xmath317 is at most countable .
secondly , we show assertion ( ) . suppose that @xmath320 , then from proposition [ mce ] , @xmath326_\beta , [ v]_{\alpha } ) \to 0 $ ] as @xmath327 .
thus , given @xmath197 , we can find a @xmath232 such that @xmath328 for all @xmath329 with @xmath330 .
so @xmath331_\alpha , v_{\alpha } ) \leq h^*([v_n]_{q_1 } , v_{\alpha } ) \leq h^*([v_n]_{q_1 } , u_{q_1 } ) + \varepsilon\ ] ] for @xmath332 .
hence , by and the arbitrariness of @xmath246 , we obtain @xmath333_\alpha , v_{\alpha } ) \to 0 \
( n\to \infty).\ ] ] on the other hand , @xmath334_{\alpha } ) \leq h^*(v_{\alpha } , [ v_n]_{q_2 } ) \leq h^ * ( u_{q_2 } , [ v_n]_{q_2 } ) + \varepsilon\ ] ] for @xmath335 .
hence , by and the arbitrariness of @xmath246 , we obtain @xmath336_\alpha ) \to 0 \
( n\to \infty).\ ] ] combined with and , we thus obtain .
* step 2 . * prove that @xmath337_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \to 0.\ ] ] given @xmath197 .
it can be deduced that , for all @xmath338 , @xmath339_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \nonumber \\
\leq & \left ( \int_0^h h ( [ v_n]_\alpha , [ v_n]_{\alpha+h } ) ^p \;d\alpha \right)^{1/p } + \left ( \int_0^h h ( [ v_n]_{\alpha+h } , [ v]_{\alpha+h } ) ^p \;d\alpha \right)^{1/p } + \left ( \int_0^h h ( [ v]_{\alpha+h } , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \nonumber \\ = & \left ( \int_h^{2h } h ( [ v_n]_{\beta - h } , [ v_n]_{\beta } ) ^p \;d\beta \right)^{1/p } + \left ( \int_h^{2h } h ( [ v_n]_{\beta } , [ v]_{\beta } ) ^p \;d\beta \right)^{1/p } + \left ( \int_h^{2h } h ( [ v]_{\beta } , [ v]_{\beta - h } ) ^p \;d\beta \right)^{1/p } \nonumber \\ \leq & \left ( \int_h^{1 } h ( [ v_n]_{\beta - h } , [ v_n]_{\beta } ) ^p \;d\beta \right)^{1/p } + \left ( \int_h^{1 } h ( [ v_n]_{\beta } , [ v]_{\beta } ) ^p \;d\beta \right)^{1/p } + \left
( \int_h^{1 } h ( [ v]_{\beta } , [ v]_{\beta - h } ) ^p \;d\beta \right)^{1/p}.\label{ean}\end{aligned}\ ] ] since @xmath40 is @xmath11-mean equi - left - continuous , there exists an @xmath340 such that @xmath341_{\beta - h } , [ v_n]_{\beta } ) ^p \;d\beta \right)^{1/p } < \varepsilon/4\ ] ] for all @xmath67 . from , we know if @xmath69 then @xmath342_{\beta - h } , [ v_n]_{\beta } ) \to h ( [ v]_{\beta - h } , [ v]_{\beta } ) $ ] a.e . on @xmath343
so , by fatou lemma , we have @xmath344_{\beta - h } , [ v]_{\beta } ) ^p \;d\beta \right)^{1/p } \leq \liminf_n \left ( \int_h^{1 } h ( [ v_n]_{\beta - h } , [ v_n]_{\beta } ) ^p \;d\beta \right)^{1/p } \leq \varepsilon/4,\ ] ] note that @xmath345_h$ ] and @xmath346_h$ ] are contained in @xmath207_\alpha : u\in u\}$ ] which is compact , it thus follows from the lebesgue s dominated convergence theorem and that @xmath347_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \to 0\ ] ] as @xmath69 .
hence there is an @xmath348 such that @xmath349_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \leq \varepsilon/4\ ] ] for all @xmath79 . combined with , , , and
, it yields that @xmath350_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \\ & \leq \left ( \int_0^h h ( [ v_n]_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } + \left ( \int_h^1 h ( [ v_n]_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \\ & \leq \varepsilon/4 + \left ( \int_h^1 h ( [ v_n]_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } + \varepsilon/4 + \left ( \int_h^1 h ( [ v_n]_\alpha , [ v]_\alpha ) ^p \;d\alpha \right)^{1/p } \\ & \leq \varepsilon\end{aligned}\ ] ] for all @xmath79 . thus we obtain from the arbitrariness of @xmath246 .
* step 3 . * show that @xmath351 . by , we know that there is an @xmath77 such that @xmath352_\alpha , [ v_n]_\alpha ) ^p \;d\alpha \right)^{1/p } < 1,\ ] ] and then @xmath353_\alpha , \{0\ } ) ^p \;d\alpha \right)^{1/p } \\ & \leq \left ( \int_0 ^ 1 h ( [ v]_\alpha , [ v_n]_\alpha ) ^p \;d\alpha \right)^{1/p } + \left ( \int_0 ^ 1 h ( [ v_n]_\alpha , \{0\ } ) ^p \;d\alpha \right)^{1/p } \\ & \leq 1 + \left ( \int_0 ^ 1 h ( [ v_n]_\alpha , \{0\ } ) ^p \;d\alpha \right)^{1/p } < + \infty.\end{aligned}\ ] ] by properties ( 1),(2 ) and ( 3 ) of @xmath304 and theorem [ rfbp ] , this yields that @xmath351 . from steps 1 , 2 and 3 , we know that for arbitrary sequence @xmath281 of @xmath40 , there exists a subsequence @xmath282 of @xmath281 which converges to @xmath351 .
this means that @xmath40 is a relatively compact set in @xmath190 .
@xmath95 [ crm ] from theorem [ pcn ] , we can obtain the following conclusion .
suppose that @xmath354 and that @xmath355 .
let @xmath356 be a decreasing sequence in @xmath216 $ ] converging to zero .
if @xmath357 satisfies conditions ( 1 ) and ( 2 ) in theorem [ pcn ] .
then the following statements are equivalent .
+ ( 1 ) @xmath358 for all @xmath215 , @xmath359 .
+ ( 2 ) @xmath360_\alpha , [ u]_\alpha ) { \stackrel } { \mbox{a.e . } } { \to } 0 \ ( [ 0,1 ] ) $ ] .
+ ( 3 ) @xmath361 and @xmath196 .
the readers can see @xcite for study on this subject , which consider the relations among @xmath4 convergence and other types of convergence on fuzzy sets spaces .
[ tcn ] @xmath40 is a totally bounded set in @xmath271 if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous . *
proof * notice that it only use the totally boundedness of @xmath40 to show the necessity part of the proof of theorem [ pcn ] .
so the desired conclusion follows immediately from theorem [ pcn ] . @xmath95 [ gscn ]
let @xmath40 be a closed subset of @xmath362 , then @xmath40 is compact in @xmath362 if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous .
* proof * the desired result follows immediately from theorem [ pcn ] .
in this section , we show that @xmath190 is the completion of @xmath363 , and present characterizations of totally bounded sets , relatively compact sets and compact sets in @xmath363 .
diamond and kloeden @xcite pointed out that @xmath6 is not a complete space .
ma @xcite gave the following example to show this fact .
let @xmath364 ma pointed out that @xmath365 is a cauchy sequence in @xmath366 , but has no limit in @xmath367 .
put @xmath368 then it can be checked that @xmath369 and @xmath66 converges to @xmath65 in @xmath370 .
notice that @xmath371 and @xmath372 , it yields that both @xmath373 and @xmath374 are not complete . along this line
, it can be shown that @xmath6 , @xmath10 , @xmath13 and @xmath363 are not complete .
[ scp ] @xmath190 is a complete space . * proof * it suffices to prove that each cauchy sequence has a limit in @xmath190 .
let @xmath375 be a cauchy sequence in @xmath190 , we assert that @xmath375 is a relatively compact set in @xmath190 . to show this assertion , by theorem [ pcn ] , it only need to prove that @xmath375 is a bounded set in @xmath190 and that @xmath375 is @xmath11-mean equi - left - continuous .
the former follows immediately from the fact that @xmath375 is a cauchy sequence .
now we prove the latter .
given @xmath197 , since @xmath376 is a cauchy sequence , there exists an @xmath377 satisfies that @xmath378 for all @xmath379 . by lemma [ uplc ] ,
@xmath380 is @xmath11-mean equi - left - continuous , hence we can find an @xmath212 such that @xmath381_{{\alpha}-h } , [ u_k]_{{\alpha } } ) ^p \;d{\alpha}\right)^{1/p } \leq \varepsilon/3 \label{nevc}\ ] ] for all @xmath382 . if @xmath383 , then @xmath384_{{\alpha}-h } , [ u_k]_{{\alpha } } ) ^p \;d{\alpha}\right)^{1/p } \nonumber \\ & \hspace{-6mm}\leq \left ( \int_h^{1 } h ( [ u_k]_{{\alpha}-h } , [ u_n]_{{\alpha}-h } ) ^p \;d{\alpha}\right)^{1/p } + \left ( \int_h^{1 } h ( [ u_n]_{{\alpha}-h } , [ u_n]_{{\alpha } } ) ^p \;d{\alpha}\right)^{1/p } + \left ( \int_h^{1 } h ( [ u_n]_{{\alpha } } , [ u_k]_{{\alpha } } ) ^p \;d{\alpha}\right)^{1/p } \nonumber \\ & \hspace{-6 mm } \leq \varepsilon/3 + \varepsilon/3 + \varepsilon/3 = \varepsilon . \label{knvc}\end{aligned}\ ] ] from the arbitrariness of @xmath246 and ineqs . and , we know @xmath385 is @xmath11-mean equi - left - continuous
. now , from the relatively compactness of @xmath385 in @xmath190 , there exists a subsequence @xmath386 of @xmath385 such that @xmath387 .
note that @xmath385 is a cauchy sequence , we thus know that @xmath66 , @xmath67 , also converges to @xmath65 in @xmath388 .
the proof is completed .
@xmath95 by theorems [ pcn ] and [ tcn ] , a set @xmath40 in @xmath190 is totally bounded if and only if it is relatively compact .
we can also deduce that @xmath190 is complete from this fact .
[ sln ] @xmath389 is a dense set in @xmath190 . * proof * given @xmath196 .
put @xmath390 , @xmath67 . then @xmath391_{\alpha}=\left\ { \begin{array}{ll } [ u]_{\alpha } , & \hbox{if } \ { \alpha}\geq 1/n , \\ \mbox{}[u]_{1/n } , & \hbox{if } \
\alpha \leq 1/n , \end{array } \right.\ ] ] for all @xmath392 $ ] . by lemma [ ulce ] , we know @xmath393_0 = [ u]_{1/n}\in k(\mathbb{r}^m)$ ] .
it thus follows from theorems [ rfbe ] and [ rfbp ] that @xmath394 for @xmath67 . since @xmath196 , we know @xmath395_{{\alpha } } ,
\{0 \ } ) ^p \;d{\alpha}\right)^{1/p }
< + \infty$ ] , thus , by the absolute continuity of the lebesgue s integral , it holds that , for each @xmath197 , there is a @xmath396 such that @xmath397_{{\alpha } } , \{0 \ } ) ^p \;d{\alpha}\right)^{1/p } < \varepsilon.\ ] ] note that @xmath398_{1/n } , [ u]_{{\alpha } } ) ^p \;d{\alpha}\right)^{1/p } \\ & \leq \left ( \int_0^{1/n } h ( [ u]_{{\alpha } } , \{0\ } ) ^p \;d{\alpha}\right)^{1/p } + \left ( \int_0^{1/n } h ( [ u]_{1/n } , \{0\ } ) ^p \;d{\alpha}\right)^{1/p } \\ & \leq 2 \left ( \int_0^{1/n } h ( [ u]_{{\alpha } } , \{0\ } ) ^p \;d{\alpha}\right)^{1/p},\end{aligned}\ ] ] it then follows from ineq .
that @xmath399 as @xmath69 .
so , for each @xmath196 , we can find a sequence @xmath400 such that @xmath401 converges to @xmath402 .
this means that @xmath389 is dense in @xmath175 .
@xmath95 [ undg ] from the proof of theorem [ sln ] , we know the following fact . + given @xmath196 , then @xmath403 for each @xmath404 , and @xmath405 as @xmath69 . from theorems [ scp ] and [ sln ]
, we get the following theorem .
@xmath406 is the completion of @xmath407 .
let @xmath408 , then @xmath40 is totally bounded in @xmath409 if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous . * proof * note that @xmath410 , we thus know @xmath411 is a totally bounded set in @xmath412 if and only if @xmath40 is a totally bounded set in @xmath406 .
so the desired conclusion follows immediately from theorem [ tcn ] .
@xmath95 [ fbchra ] let @xmath408 , then @xmath40 is compact in @xmath409 if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) @xmath413 , where @xmath414 is the closure of @xmath40 in @xmath190 . * proof * the desired result follows immediately from theorem [ gscn ] .
@xmath95 condition ( 3 ) in theorem [ fbchra ] involves the closure of @xmath40 in the completion space @xmath190 .
next , we look for an characterization of compactness that only depends on @xmath40 itself . the following concept is needed .
let @xmath415 and @xmath416 be the characteristic function of @xmath417 , where @xmath418 is a positive real number .
given @xmath419 , then @xmath420 .
define @xmath421_{\alpha } , [ \widehat{b_r } ] _ { \alpha } ) ^p \ , d{\alpha}\right ) ^ { 1/p}.\ ] ] it can be checked that , for @xmath419 , @xmath422 if and only if @xmath104_0
\subseteq b_r $ ] .
note that @xmath423_{\alpha } , [ v \vee \widehat{b_r } ] _ { \alpha } ) \leq h ( [ u]_{\alpha } , [ v]_{\alpha } ) , \ ] ] it thus holds that @xmath424 [ fbsrchra ] let @xmath408 , then @xmath40 is relatively compact in @xmath409 if and only if @xmath40 satisfies conditions ( 1 ) , ( 2 ) in theorem [ pcn ] and the following condition ( 3@xmath425 ) . ( 3@xmath425 ) : : given @xmath426 , there exists a @xmath427 and a subsequence @xmath282 of @xmath281 such that @xmath428 . * proof * suppose that @xmath40 is a relatively compact set but does not satisfy condition ( 3@xmath425 ) .
take @xmath429 , then there exists @xmath430 and a subsequence @xmath431 of @xmath432 such that @xmath433 for all @xmath67 .
if @xmath291 and positive numbers @xmath434 have been chosen , we can find a subsequence @xmath292 of @xmath293 and @xmath435 such that @xmath436 for all @xmath67 .
put @xmath299 for @xmath67 .
then @xmath282 is a subsequence of @xmath281 and @xmath437 for @xmath301 .
let @xmath438 be a accumulation point of @xmath282 .
it then follows from and that @xmath439 for all @xmath301 .
so we know @xmath440 .
this contradicts the fact that @xmath40 is a relatively compact set in @xmath441 .
suppose that @xmath442 satisfies condition ( 3@xmath425 ) .
given a sequence @xmath281 in @xmath40 with @xmath443 .
then , from , there exists a @xmath427 such that @xmath444 . hence @xmath104_0 \subseteq b_r $ ] , i.e. @xmath445 .
so , by theorem [ pcn ] , we know that if @xmath40 meets conditions ( 1 ) , ( 2 ) and ( 3@xmath425 ) , then @xmath40 is a relatively compact set in @xmath446
. @xmath95 [ fbschra ] let @xmath40 be a closed set in @xmath447 , then @xmath40 is compact in @xmath409 if and only if @xmath40 satisfies conditions ( 1 ) , ( 2 ) and ( 3@xmath425 ) in theorem [ fbsrchra ] .
* proof * the desired result follows immediately from theorem [ fbsrchra ] .
in this section , we consider relationship among all fuzzy sets spaces mentioned in this paper .
they are subspaces of @xmath409 or @xmath362 .
the conclusions are summarized in fig.[sbpren ] . by using these facts and the results in sections [ crfpat ] and [ relation ]
, we gives characterizations of totally bounded sets , relatively compact sets and compact sets in these subspaces .
[ gsmc ] @xmath137 is a closed set in @xmath362 * proof * it only need to show that each accumulation point of @xmath137 belongs to itself . given a sequence @xmath281 in @xmath137 with @xmath448 , then clearly @xmath254_\alpha , [ u]_\alpha ) { \stackrel } { \mbox{a.e . } } { \to } 0 \ ( [ 0,1 ] ) $ ] .
suppose that @xmath144 $ ] .
if @xmath254_\alpha , [ u]_\alpha ) \to 0 $ ] , then by theorem [ ksc ] , @xmath104_{\alpha}\in k_s(\mathbb{r}^m)$ ] . if @xmath254_\alpha , [ u]_\alpha ) \not\to 0 $ ] , then there exists a sequence @xmath449 such that @xmath450_{\beta_n } \in k_s(\mathbb{r}^m)$ ] .
note that @xmath104_{\alpha}= \bigcap_{n } [ u]_{\beta_n } $ ] , this implies that @xmath450_\alpha \in k_s(\mathbb{r}^m)$ ] .
so we know @xmath451
. @xmath95 [ slne ] @xmath12 is a dense set in @xmath23 . * proof * given @xmath165 . put @xmath390 , @xmath67
. then @xmath452 . from remark [ undg ] , we know that @xmath399 as @xmath69 .
so @xmath12 is dense in @xmath23 .
@xmath95 [ sgmcde ] @xmath136 is a closed subset of @xmath23 . * proof * to show that @xmath136 is a closed set in @xmath453 , let @xmath281 be a sequence in @xmath136 which converges to @xmath451
, we only need to prove that @xmath167 . since @xmath454_{\alpha } , [ u]_{\alpha})^p \ ; d\alpha \right)^{1/p } \to 0 $ ] , it holds that @xmath455_{\alpha } , [ u]_{\alpha } ) \to 0 \ \hbox{a.e . on } \ [ 0,1].\ ] ] hence @xmath456 is a bounded set in @xmath208 , and therefore @xmath457 we assert that @xmath458_{\alpha}\ \hbox{for all } \ { \alpha}\in ( 0,1].\ ] ]
so , from and , we know @xmath459 } \mbox{ker}\ ; [ u]_{\alpha}= \mbox{ker}\ ; u.\ ] ] it thus follows from theorem [ rs0p ] that @xmath167 .
now we prove .
the proof is divided into two cases .
+ * case 1*. @xmath160 $ ] satisfies that @xmath254_{\alpha } , [ u]_{\alpha } ) \to 0 $ ] .
+ in this case , by corollary [ kef ] , we have that @xmath460_{\alpha}\subset \mbox{ker}\ ; [ u]_{\alpha}.\ ] ] * case 2*. @xmath160 $ ] satisfies that @xmath254_{\alpha } , [ u]_{\alpha } ) \not\to 0 $ ] .
+ by , we know that there is a sequence @xmath461 $ ] , @xmath67 , such that @xmath462 and @xmath254_{{\alpha}_n } , [ u]_{{\alpha}_n } ) \to 0 $ ] . from case 1 , we obtain that @xmath463_{{\alpha}_n}.\ ] ] note that @xmath464_{\alpha } , [ u]_{{\alpha}_n } ) \to 0 $ ] , so , by corollary [ kef ] , @xmath465_{{\alpha}_n } \subset \mbox{ker}\ ; [ u]_{{\alpha}},\ ] ] combined and , we get that @xmath466_{{\alpha}}. \quad \box\ ] ] [ ukc ] suppose that @xmath281 is a sequence in @xmath22 and that @xmath165 . if @xmath399 , then @xmath167 and @xmath467 .
* proof * the desired result follows immediately from the proof of theorem [ sgmcde ] .
@xmath95 @xmath10 is a closed subspace of @xmath13 .
* proof * the desired result follows immediately from corollary [ ukc ] .
@xmath95 [ sdln ] @xmath7 is a dense set in @xmath22 . * proof * the proof is similar to the proof of theorem [ sln ] . @xmath95
[ vmce ] @xmath21 is a closed subspace of @xmath22 .
* proof * by proposition [ kcs ] , we know that @xmath468 is a closed set in @xmath469 . in a way similar to the proof of theorem [ sgmcde ] , we can obtain the desired result by using this fact .
@xmath95 @xmath6 is a closed subspace of @xmath10 . *
proof * the desired results follows immediately from theorem [ vmce ] .
@xmath95 [ edln ] @xmath5 is a dense set in @xmath21 . * proof * the proof is similar to the proof of theorem [ sdln ] . @xmath95
[ gsc ] @xmath470 is the completion of @xmath471 . * proof * the desired result follows from theorems [ scp ] , [ gsmc ] , [ slne ] .
@xmath95 [ smc ] @xmath22 is the completion of @xmath10 . * proof * the desired result follows from theorems [ sgmcde ] , [ sdln ] , [ gsc ] . @xmath95
[ evmc ] @xmath21 is the completion of @xmath6 . * proof * the desired result follows from theorems [ vmce ] , [ edln ] , [ smc ] .
@xmath95 we use @xmath472 to denote the closure of @xmath473 in @xmath22 . from corollary [ ukc ] and remark [ undg ] , we know that 1 . @xmath474 .
2 . @xmath9 is a closed subset of @xmath10 .
3 . @xmath475 is a closed subset of @xmath22 . .
@xmath476 denotes that @xmath477 is a closed subspace of @xmath478 and @xmath479 means that @xmath478 is the completion of @xmath477 .
, title="fig : " ] + we summarize above facts in figure [ sbpren ] . based on these facts
, we will discuss characterizations of totally bounded sets , relatively compact sets and compact sets in various subspaces of @xmath480 in the sequel .
[ spcn ] let @xmath481 ( @xmath482 , @xmath483 , @xmath484 ) , then @xmath40 is totally bounded if and only if it is relatively compact in @xmath485 ( @xmath21 , @xmath486 , @xmath487 ) , which is equivalent to + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded , and + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous .
* proof * note that in a complete space , a set is totally bounded is equivalent to it is relatively compact .
so the desired results follow from theorems [ pcn ] and the completeness of @xmath485 , @xmath21 , @xmath486 and @xmath487 .
@xmath95 [ smpcn ] let @xmath40 be a closed set in @xmath485 ( @xmath21 , @xmath486 , @xmath487 ) , then @xmath40 is compact in @xmath485 ( @xmath21 , @xmath486 , @xmath487 ) if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous . *
proof * the desired results follow immediately from theorem [ spcn ] .
@xmath95 let @xmath488 ( @xmath489 , @xmath490 , @xmath491 ) , then @xmath40 is totally bounded in @xmath492 ( @xmath493 , @xmath494 , @xmath495 ) if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous .
* proof * the desired conclusion follows immediately from theorem [ tcn ] .
@xmath95 [ gsmchra ] let @xmath488 ( @xmath489 , @xmath490 , @xmath491 ) , then @xmath40 is compact in @xmath492 ( @xmath496 , @xmath497 , @xmath498 ) if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) @xmath413 , where @xmath414 is the closure of @xmath40 in @xmath485 ( @xmath21 , @xmath486 , @xmath487 ) . *
proof * the desired results follow from theorem [ gscn ] and the completeness of @xmath485 , @xmath21 , @xmath486 and @xmath487 .
@xmath95 [ fbsrchrae ] let @xmath488 ( @xmath489 , @xmath490 , @xmath491 ) , then @xmath40 is relatively compact in @xmath158 ( @xmath499 , @xmath500 , @xmath10 ) if and only if + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3@xmath425 ) given @xmath426 , there exists a @xmath427 and a subsequence @xmath282 of @xmath281 such that @xmath428 .
* proof * note that @xmath501 , @xmath502 , @xmath503 , @xmath504 , so we can obtain the desired results by applying theorems [ fbsrchra ] and [ spcn ] .
@xmath95 [ fbschrae ] let @xmath40 be a closed set in @xmath505 ( @xmath506 , @xmath507 , @xmath7 ) , then @xmath40 is compact in @xmath158 ( @xmath499 , @xmath500 , @xmath10 ) if and only if @xmath40 satisfies conditions ( 1 ) , ( 2 ) and ( 3@xmath425 ) in theorem [ fbsrchrae ] .
* proof * the desired result follows immediately from theorem [ fbsrchrae ] .
@xmath95 compare theorem [ smpcn ] with proposition [ gsp ] .
we can see that , in contrast to the former , the latter has an additional condition ( 3 ) .
the reason is that @xmath10 is not complete .
the function of `` conditions ( 3 ) '' in proposition [ gsp ] is to guarantee completeness of the closed subspace @xmath508 of @xmath10 .
in fact , under the assumptions of proposition [ gsp ] , by remark [ crm ] , we know that if @xmath217 converges to @xmath223 in @xmath4 metric , then there exists a @xmath509 such that @xmath510 and @xmath511_{\alpha}= [ u_0^{(r_i)}]_{\alpha}= [ u(r_i)]_{\alpha}$ ] when @xmath220 .
so proposition [ gsp ] can also be written as : * proposition [ gsp]@xmath425 * a closed set @xmath40 in @xmath512 is compact if and only if : + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) @xmath513 , where @xmath514 is the closure of @xmath40 in @xmath22 . similarly , propositions [ gs0p ] and [ gep ] can also be written as : * proposition [ gs0p]@xmath425 * a closed set @xmath40 in @xmath8 is compact if and only if : + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) @xmath515 , where @xmath514 is the closure of @xmath40 in @xmath516 .
* proposition [ gep]@xmath425 * a closed set @xmath517 is compact if and only if : + ( 1 ) @xmath40 is uniformly @xmath11-mean bounded ; + ( 2 ) @xmath40 is @xmath11-mean equi - left - continuous ; + ( 3 ) @xmath518 , where @xmath514 is the closure of @xmath40 in @xmath21 .
the authors wish to thank the anonymous reviewers for their invaluable comments and suggestions . | diamond gave compact criteria in fuzzy numbers space endowed with @xmath0 metric and compact criteria in the space of fuzzy star - shaped numbers with respect to the origin endowed with @xmath0 metric .
however , these compact criteria are wrong . wu and zhao proposed right characterizations in these two spaces . based on this result , zhao and wu further gave compact criteria in the space of fuzzy star - shaped numbers with @xmath0 metric .
however , compare the existing compactness characterizations of fuzzy sets spaces endowed with @xmath0 metric with arzel
ascoli theorem , it finds that the latter gives the compact criteria by characterizing the totally bounded sets while the former does not seem to characterize the totally bounded sets . since , in metric spaces , totally boundedness is a key feature of compactness .
we present characterizations of totally bounded sets , relatively compact sets and compact sets in the fuzzy sets spaces @xmath1 and @xmath2 equipped with @xmath0 metric , where @xmath1 and @xmath2 are two kinds of fuzzy sets on @xmath3 which do not have any assumptions of convexity or star - shapedness .
all fuzzy sets spaces considered in this paper are subspaces of @xmath2 endowed with @xmath0 metric . based on these characterizations and the discussions on convexity and star - shapedness of fuzzy sets ,
we construct the completions of every fuzzy sets space mentioned in this paper
. then we clarify relation among all the ten fuzzy sets spaces discussed in this paper including the general fuzzy star - shaped numbers space introduced by qiu et al . at last
, it gives characterizations of totally bounded sets , relatively compact sets and compact sets in all the fuzzy sets spaces mentioned in this paper .
[ section ] [ section ] [ section ] [ section ] [ section ] [ section ] [ section ] fuzzy sets ; compact sets ; totally bounded sets ; @xmath0 metric ; star - shaped sets |
globular clusters ( gcs ) are simple stellar populations ( single age , single metallicity ) that help to trace the formation history of their host galaxy .
gcs are also powerful tools in revealing the halo kinematics and dynamics , if large samples of globular cluster radial velocities can be obtained .
ngc 5128 , the central galaxy in the centaurus cluster 3.9 mpc away , and the nearest readily accessible giant e galaxy , provides a unique and exciting possibility to study the dynamics and formation history through its globular cluster system ( gcs ) .
however , finding its clusters individually is a difficult job : at its moderately low galactic latitude of @xmath3 , field contamination by both foreground stars and background galaxies is significant , and the gcs can not be distinguished by only image morphology , object size , or color .
thus except for the very expensive and time - consuming approach of resolving the gcs into individual stars by ( say ) hst imaging , radial velocity determinations provide the only definitive method for classifying a candidate object in the ngc 5128 field as a genuine gc .
ngc 5128 has a statistically determined total population of @xmath4 globular clusters , giving it a low specific frequency @xmath5 ( harris , harris , & geisler 2004 , hereafter hhg04 ) .
an @xmath6 value this low ( in the range occupied by large spiral galaxies ) is suggestive of a merger origin @xcite ; other indicators of mergers are the faint isophotal shells in the halo of the galaxy ( e.g. * ? ? ?
* ) , and the activity in the inner 5 kpc region indicated by the visible dust lane , the recent star formation , and the presence of gas .
the gcs , with its wide mixture of cluster metallicities and perhaps ages , should provide many additional clues to the formation history surrounding this giant elliptical . an excellent starting point for
any such discussion is to have as large as possible a list of individual , certainly identified gcs through radial velocity measurements .
previous work has led to a list of 215 confirmed gcs in ngc 5128 with radial velocity measurements ( peng , ford , & freeman 2004 , hereafter pff04 ) .
many hundreds more clusters remain to be found . in this paper ,
the first of a new series of studies of the gcs in this unique galaxy , we present new velocity measurements and the discovery of 31 new gcs .
the wide - field photometry study of the ngc 5128 field in the washington @xmath7 system by @xcite provides a list of 327 gc candidates having globular - cluster - like colors and slightly nonstellar morphology .
this list provides the basis for our new radial velocity survey .
observations were taken at the cerro tololo inter - american observatory ( ctio ) 4 m telescope in april 2004 . the hydra multi - object fibre - fed spectrograph , with @xmath8 diameter field of view ,
was used with the kpgl1 grating and @xmath9 binning in dispersion , giving a spectral resolution element of approximately 3.5 binned pixels ( fwhm ) .
the spectral coverage was @xmath10 @xmath11 @xmath12 , extending from ca ii h+k at the blue end to the mgb features at the red end at 1.18 / px . of our three assigned nights ,
two were lost to clouds , and the third had poor seeing throughout ( @xmath13 ) . nevertheless , we successfully observed two hydra field configurations , both centered on the galaxy .
total exposure times were @xmath14 sec for the first field and @xmath15 sec for the second .
the data reduction was performed through the _ dohydra _ package within iraf , following the normal sequence of preprocessing described on the hydra webpages .
the raw images were overscan and bias subtracted , and the detector was flat - fielded with milk flats ( daytime sky exposures ) using a diffusing filter in front of the hydra fibres .
cosmic rays were subtracted with a laplacian edge detection method @xcite .
spectral throughput and dispersion corrections were applied based on the standard continuum lamp and hydra henear penray lamp to obtain wavelength calibrated spectra .
removal of background light from each object spectrum includes both the uniform dark - sky illumination and ( for the gc candidates close to the galaxy center ) the bulge light from ngc 5128 itself . to trace both effects
, we placed 40 - 50 fibers in each hydra field configuration on blank - sky positions sprinkled across the entire field , then plotted the mean intensity of each sky fiber versus radius from galaxy center .
the data showed that the bulge light profile was well matched by a @xmath16 radial dependence ( see , e.g. * ? ? ? * ) .
for radii larger than @xmath17 , the galaxy bulge light contamination was negligible and we simply subtracted the average of the sky spectra at these larger radii from all the object spectra . for the inner objects , we then removed the additional bulge light contamination through appropriate radial scaling of the average bulge sky spectrum . from the fully preprocessed and background - subtracted spectra , we derived radial velocities obtained through fourier cross - correlation , using the _ fxcor _ task in the _ radial velocity _ iraf package .
the spectrum of each candidate object was correlated against a template spectrum of known velocity and the correlation peak was fit with a lorentzian function .
our template spectrum was constructed from the sum of five of the previously known bright gcs in ngc 5128 ( c7 , c11 , c17 , c19 , and c20 ) , which we observed in both of our hydra fields and which have well determined velocities taken from pff04 .
thus , the template was very well tuned to the the spectral features in the gcs that we are searching for since it was built from exactly the same type of object .
ngc 5128 has a systemic velocity of 541 km s@xmath18 @xcite and an internal stellar velocity dispersion of @xmath19 km s@xmath18 @xcite .
these features allow a clean velocity separation between gcs and foreground stars in the milky way or background galaxies .
as shown ( e.g. ) by pff04 , the foreground stars have velocities almost all less than @xmath20 km s@xmath18 , while any faint background galaxies will have velocities of many thousands of km s@xmath18 . any objects that fall in the range @xmath21 km s@xmath18 ( and drawn from the candidate list pre - selected by image morphology , magnitude and color ) will therefore be near - certain gcs . from both hydra fields combined , we obtained radial velocity measurements for 86 candidate gcs and 43 previously known gcs .
the previously known clusters were used to help establish the zero point of our velocity scale and check the internal measurement uncertainties ( see below ) . of the candidates , we found 31 with @xmath22 km s@xmath18 , which we identify as newly discovered gcs .
of these 31 , 5 are x - ray sources previously detected by @xcite . in table
[ newgc ] , we list the data for these gcs .
successive columns give the identification number ( whh prefixes denote our new objects ) , coordinates , radial distance from the center of ngc 5128 , @xmath23 and @xmath0 photometry from hhg04 , radial velocity , and velocity uncertainty from the _ fxcor _ solution . of the 55 remaining candidates , 24 are definite foreground stars with @xmath24 kms@xmath18 , and 31 are likely background galaxies ;
these are listed in table [ fsbg ] .
the objects with null results for the radial velocity are ones whose spectra had a poor correlation with the gc template spectrum , suggestive of a background galaxy with either very large redshift or spectral features very different from the template . confirming 31 candidates as gcs out of the 86 candidates observed yields an encouraging 36% hit rate given the extremely high field contamination level , and demonstrates the value of rigorous preselection by color and object morphology ( see the discussion of hhg04 ) .
table [ knowngc ] lists the resulting radial velocity measurements of the 43 previously known gcs remeasured in this study .
the identification number prefixes c , g , or pff_gc are originally listed by @xcite and pff04 .
figure [ positions ] shows the locations of the newly confirmed and previously known gcs .
whereas most previous surveys have concentrated on fields along the isophotal major axis of ngc 5128 ( along a line running sw to ne ) , it is worth noting that many of our newly identified gcs lie along the less well studied minor axis parallel to the inner dust lane .
there are likely to be many more to be found still .
figure [ fsgc ] shows the velocity distribution of the gcs and foreground stars measured from this study ; the two types of objects are clearly distinguished by radial velocity at @xmath25 km s@xmath18 .
the analysis of the gcs radial profile by hhg04 showed that the majority of gcs lie within @xmath26 from the galaxy center , with the main contribution of the clusters residing in an even more compact environment within @xmath27 .
our new sample of 31 gcs has a median value of @xmath28 .
the mean radial velocity difference between the previously known clusters in our study and those of pff04 is less than 10 km s@xmath18 , well within the standard deviation of the mean ( see below ) .
the zero point of our velocity scale therefore needs no adjustment .
the raw spectra of clusters whh-12 , whh-25 , whh-31 , pff_gc-073 , and the non - gc candidates 100 , 133 , and 136 had a few intensity spikes that could not be removed in the preprocessing stage .
these regions were artificially removed before the cross - correlation solutions , resulting in higher than normal velocity uncertainties ( see tables [ newgc ] & [ fsbg ] ) . for the previously known cluster c10
, we measure @xmath29 km s@xmath18 .
this differs quite a bit from the pff04 measurement of @xmath30 km s@xmath18 , but it agrees well with @xcite ( @xmath31 km s@xmath18 ) and also with the new measurement from @xcite ( 836 km s@xmath18 ) .
we therefore believe our measurement to be in the correct range .
there is one candidate for which the velocity solution revealed a second correlation peak only slightly smaller than the highest peak that is automatically found and fit by the _
fxcor _ task . for whh-22 ,
the highest correlation peak yields @xmath32 km s@xmath18 , clearly classifying this object as a gc ; but a second peak yields @xmath33 km s@xmath18 , which would classify it as a foreground star .
visual inspection of the candidate spectrum shows hydrogen and calcium line strengths consistent with those for clusters .
we keep this object as a probably gc , but with these uncertainties in mind .
a strong secondary peak was also revealed for the known cluster , pff_gc-060 .
the highest peak gives @xmath34 km s@xmath18 , but a second peak gives 217 km s@xmath18 , which would put it on the borderline between gcs and foreground stars .
its spectrum does not have high enough signal - to - noise to determine upon visual inspection whether or not it is a cluster .
however , this object has been measured by pff04 to have @xmath35 km s@xmath18 , agreeing with our highest peak correlation value . for three previously known gcs , c21 , pff_gc-089 , and pff_gc-064 , the highest correlation peak fit generated by _ fxcor _ yielded radial velocities that were inconsistent with previous measurements
. a closer look at the correlated spectra shows a second strong correlation peak in all cases . choosing the secondary peak gave @xmath36 km s@xmath18 for c21 ( compared with @xmath37 km s@xmath18 from pff04 ) and @xmath38 km s@xmath18 for pff_gc-089 ( compared with @xmath39 km s@xmath18 from pff04 ) .
c21 has also been previously measured by @xcite to be at @xmath40 km s@xmath18 .
for these two objects we therefore adopted the second correlation peaks . the known cluster , pff_gc-064 , measured in both field configurations , presents a slightly different problem : the initial poorly correlated fit in field 1 nominally classified it as a possible background galaxy .
yet , there is a second strong correlation peak with @xmath41 km s@xmath18 , agreeing with the field 2 result for the same object of @xmath42 km s@xmath18 and also with the pff04 measurement of @xmath43 km s@xmath18 .
there are three other candidates measured in both field configurations , 43/44 , 109/110 , and 101 .
their radial velocities classified these objects as possible background galaxies , with their averaged radial velocities as displayed in tables [ fsbg ]
. objects 43 and 44 in the candidate list of hhg04 turn out to be the same object , and similarly candidates 109 and 110 are the same .
lastly , we note that 12 of the hhg04 candidate objects measured in this study turned out to be previously known gcs . these objects are listed in table [ overlaps ] .
the radial velocity uncertainties we quote in the tables are from the _ fxcor _ solutions .
generally , these are well correlated with the object s apparent magnitude and thus the signal - to - noise of the spectra .
figure [ resultserrmag3 ] shows the nominal radial velocity uncertainties plotted versus v magnitude .
the new gcs ( excluding whh-12 , whh-25 , and whh-31 ; see above ) with @xmath44 had a mean @xmath45 km s@xmath18 and s / n @xmath46 per pixel in the continuum region near @xmath47 . for @xmath48 ,
the mean uncertainty was @xmath49 km s@xmath18 and s / n @xmath50 per pixel .
we tested the _
fxcor _ uncertainties in several ways .
first , the comparison between our velocities and those of pff04 for 42 clusters in common ( excluding c10 as mentioned above ) is shown in figure [ comparison ] .
the standard deviation of the differences between the two studies was @xmath5144.5 km s@xmath18 .
if the two studies have comparable internal uncertainties , then this suggests that for gcs brighter than @xmath52 the typical uncertainty of our measurements is no larger than @xmath53 km s@xmath18 , in agreement with the _ mknoise _ simulations in fig .
[ resultserrmag3 ] ( see below ) .
another internal test was to compare the velocities for objects that we measured in both hydra field settings .
there were 13 such gcs measured twice , 12 of them brighter than @xmath54 : c2 , c6 , c13 , c14 , c18 , c19 , c20 , c26 , c37 , c38 , c44 , pff_gc-056 , and pff_gc-064 ( @xmath55 mag ) .
for these , the mean velocity difference between the two fields was 3.9 km s@xmath18 , with a standard deviation of @xmath5122.3 km s@xmath18 , again consistent with fig .
[ resultserrmag3 ] . finally , we compared the _ fxcor _ uncertainties with those generated from simulated noisy spectra , adding gaussian noise to the template spectrum with the iraf _
artdata.mknoise _ task .
several dozen of these artificial spectra were cross - correlated against the original template spectrum and the resulting radial velocity uncertainties plotted against the s /
n for each noisy spectrum .
the results verify that for s / n @xmath56 per pixel ( corresponding roughly to @xmath44 ) , the velocity uncertainty was smaller than @xmath57 km s@xmath18 .
fainter than this , the uncertainty rises exponentially , reaching @xmath58 km s@xmath18 at s / n @xmath59 per pixel ( or @xmath60 ) . in fig .
[ resultserrmag3 ] , the solid line shows the velocity uncertainty increasing with v magnitude , from the _ mknoise _ simulations .
this ideal line is the minimum uncertainty due only to random noise , providing a lower boundary to the observed uncertainties . in summary , we find that the results generated by _ fxcor _ ( with its uncertainities
as listed in tables [ newgc ] , [ fsbg ] , & [ knowngc ] ) provide estimates of the internal velocity uncertainties that are quite close to the estimates from simulations and from comparisons with previous data
. this may in large part be because we deliberately used a standard velocity template which accurately matches the program objects .
we conclude that the internal uncertainties of our measured velocities are near @xmath61 km s@xmath18 for the very brightest clusters , increasing to @xmath58 km s@xmath18 for the ones at @xmath60 or fainter .
a new analysis of the radial velocities and kinematics of the confirmed gcs in ngc 5128 , with a much enlarged cluster sample over that used in pff04 , will be discussed in a later paper @xcite . for the present , we provide here only a brief discussion on the metallicity distribution function ( mdf ) .
the top panel in figure [ mdf ] shows the mdf for a total of 299 gcs in ngc 5128 that are radial - velocity members _ and _ that have available @xmath0 photometric indices : 168 are previously confirmed clusters ( pff04 ) , 31 are the new clusters from this study , and 100 more are from the upcoming study of @xcite .
although there are a few dozen other highly probable gcs that have recently been found in ngc 5128 from image morphology studies @xcite , we restrict the current discussion to an almost certainly `` pure '' sample generated by the combination of object color , morphology , and radial velocity . the dereddened colors @xmath62 from hhg04 were transformed to [ fe / h ] through the conversion derived by @xcite calibrated through the milky way cluster data .
we adopt a foreground reddening value of @xmath63 for ngc 5128 , corresponding to @xmath64 . because the transformation is slightly nonlinear
, the uncertainty in [ fe / h ] is a function of metallicity . for a typical color uncertainty @xmath65 , the corresponding uncertainty in [ fe / h ] is near @xmath66 dex at the metal - poor end , decreasing to @xmath67 dex at the metal - rich end .
the mdf , binned in steps of @xmath68 } = 0.10 $ ] , is shown in figure [ mdf ] , along with the same distribution for the milky way clusters ( data from * ? ? ?
as has been found in all previous studies of ngc 5128 starting with @xcite , the mdf is extremely broad and populated by roughly equal numbers of metal - poor and metal - rich clusters .
the color bimodality of ngc 5128 has recently been confirmed by @xcite in 4 different photometric indices .
they used spectroscopic line indices of h@xmath69 versus [ mgfe]@xmath70 to suggest the blue gc population was metal - poor and old ( @xmath71 gyr ) and the red gc population was metal - rich and young ( mean age @xmath72 gyr ) .
a bimodal - gaussian distribution ( shown in fig . [ mdf ] ) provides a reasonable fit to the total mdf , but not as cleanly as in many other giant galaxies including the milky way ( plotted for comparison in the lower panel of the figure ) .
the fit of the bimodal model to the mdf is not well constrained , allowing a wide range of particular combinations of gaussians .
the two centroids shown , at @xmath1 } = -1.46 $ ] and @xmath2 , are similar to the fit obtained by hhg04 ( see their figure 10 ) from a smaller cluster sample .
the metal - poor part of our entire sample makes up @xmath73 of the total population and the metal - rich part @xmath74 , in close agreement with hhg04 .
it does not seem likely that internal differential reddening of the clusters is an important factor in creating the color spread , since reddening - generated differences in color of several tenths of a magnitude would be needed to move a cluster from the metal - poor mode to ( e.g. ) [ fe / h ] @xmath75 , yet all the objects with @xmath0 photometry lie clearly outside the dust lane and into the clear regions of the bulge and halo ( see * ? ? ? * for extensive discussion ) .
as has long been realized , the milky way mdf ( bottom panel in figure [ mdf ] ) provides a clean match to the bimodal - gaussian model , with peaks at @xmath1 } = -1.58 \pm 0.05 $ ] and @xmath76 .
the standard deviations are 0.32 and 0.23 for the metal - poor and metal - rich subpopulations , both narrower than their counterparts for ngc 5128 .
both centroids are slightly but noticeably more metal - poor than those in ngc 5128 . by comparison ,
the mdf of the m31 globular clusters shows [ fe / h ] peaks that are more similar to ngc 5128 .
@xcite find m31 centroids of @xmath1 } = -1.44 \pm 0.03 $ ] and @xmath77 .
thus , it is likely that if ngc 5128 is a product of major mergers , then it would need m31-sized galaxies as progenitors , rather than systems like the milky way which have lower - metallicity clusters .
if we assume that ngc 5128 was built through a series of mergers extending for several gigayears , then the gcs might contain a wide range of ages .
thus the mdf shown here , based on photometric metallicities from @xmath0 with the assumption that all the clusters are homogeneously old , would underestimate the true metallicity of younger clusters .
this effect would artificially increase the spread in the mdf and partially blur out a distribution that was intrinsically bimodal .
the classic age / metallicity degeneracy for broad band colors prevents us from distinguishing this alternative from an intrinsically broad , even mdf .
figure [ fehv ] displays the ngc 5128 sample divided into bright ( @xmath78 ) and faint ( @xmath79 ) clusters .
interestingly , the fainter sample appears to have a higher proportion of metal - rich clusters than the brighter sample . a direct least - squares correlation of [ fe / h ] versus v magnitude shows no significant trend within the larger scatter in [ fe / h ] .
it is possible that we are simply seeing an intrinsically broad and uniform spread of cluster metallicities , in which the younger and more metal - rich ones tend to be of somewhat lower mass or luminosity , consistent with the age differences suggested by pff04 .
a wide range of ages would further increase the apparent spread of the mdf as seen in fig .
[ fehv ] .
further spectroscopic studies and more detailed modelling will be needed to test these speculations . in later analyses
@xcite we will discuss the age distributions and kinematics of the cluster system further . in figure [ fehrad ] ,
the cluster metallicities are displayed as a function of projected galactocentric radius .
both metal - poor ( @xmath1 } > -1.0 $ ] ) and metal - rich ( @xmath1 } < -1.0 $ ] ) clusters have been found extending out to a radius of @xmath8 ( @xmath80 kpc ) .
however , beyond @xmath81 , there are double the number of metal - poor compared to metal - rich clusters .
the mean metallicity beyond this distance is @xmath1 } = -1.21 $ ] with a standard deviation of 0.47 . as is discussed in more detail in hhg04
, this change in ratios of subpopulations is not likely to be due to selection bias and is therefore probably intrinsic to the system .
we present new radial velocities for 86 globular cluster candidates selected from @xcite .
the results show 31 newly identified clusters with @xmath22 kms@xmath18 , 24 definite foreground stars with @xmath24 km s@xmath18 , and 31 probable background galaxies .
we also obtained velocities for 43 previously known gcs in ngc 5128 .
the internal uncertainty in our @xmath82 measurements ranges from @xmath61 km s@xmath18 at the bright end of the sample ( @xmath83 to @xmath84 km s@xmath18 at @xmath60 . combining our data with previously published material , we derive a metallicity distribution for the gcs based on 299 clusters .
a bimodal distribution with centroids at @xmath1 } = -1.46 $ ] and @xmath2 provides an approximate fit to the mdf , though not highly accurate , and the data hint that the relative proportions of metal - poor and metal - rich clusters may change with luminosity .
two possible interpretations are that ( a ) we may be seeing the combined effects of a several - gy internal age spread in the cluster system , combined with a trend for the younger clusters to be progressively less massive , or ( b ) the mdf is genuinely broad and evenly distributed , unlike the more cleanly bimodal mdfs seen in other giant galaxies .
we greatly appreciate the data reduction advice and support provided by knut olsen at ctio , and also advice regarding uncertainty tests from david james at vanderbilt university .
we thank mike beasley for transmitting cluster velocity data in advance of publication .
this work was supported by the natural sciences and engineering research council of canada through research grants to glhh and weh .
beasley , m. et al .
2005 , in preparation gomez , m. et al .
2005 , in preparation harris , g.l.h . ,
geisler , d. , harris , h.c .
, & hesser , j.e .
1992 , , 104 , 613 harris , g.l.h . ,
geisler , d. , harris , w.e . ,
schmidt , b.p . ,
hesser , j.e . ,
reid , m. , milne , m. , hulme s.c . , & kidd , t.t .
2004 , , 128 , 712 harris , g.l.h . ,
harris , w.e . , & geisler , d. 2004 , , 128 , 723 ( hhg04 ) harris , w.e .
1996 , , 112 , 1487 harris , w.e .
2001 , in _ star clusters _ ,
saas - fee advanced course 28 ( new york : springer ) , ed .
l. labhardt & b. binggeli harris , w.e .
& harris , g.l.h . 2002 , , 123 , 3108 harris , w.e . ,
harris , g.l.h . ,
holland , s.t .
, & mclaughlin , d.e .
2002 , , 124 , 1435 hesser , j.e . ,
harris , h.c . , & harris , g.l.h .
1986 , , 303 , l51 hui , x. , ford , h.c . , freeman , k.c . , & dopita , m.a .
1995 , , 449 , 592 kraft , r.p .
, kregenow j.m . , forman , w.r .
, jones , c. , & murray , s.s .
2001 , , 560 , 675 peng , e.w . ,
ford , h.c . , & freeman , k.c .
2004 , , 150 , 367 ( ppf04 ) peng , e.w . , ford , h.c . , & freeman , k.c .
2004 , , 602 , 705 peng , e.w . ,
ford , h.c . ,
freeman , k.c .
, & white , r.l .
2002 , , 124 , 3144 perrett , k.m . , bridges , t.j . ,
hanes , d.a . ,
irwin , m.j . , brodie , j.p .
, carter , d. , huchra , j.p . , & watson , f.g .
2002 , , 123 , 2490 rejkuba , m. 2001 , , 369 , 812 van den bergh , s. , hesser , j.e . ,
& harris , g.l.h .
1981 , , 86 , 24 van dokkum , p.g .
2001 , , 113 , 1420 wilkinson , a. , sharples , r.m . , fosbury , r.a.e . ,
& wallace , p.t . 1986 , , 218 , 297 woodley , k.a .
2005 , in preparation whh-1 & 13 24 21.40 & -43 02 36.8 & 12.19 & 18.23 & 1.484 & 600 & 35 + whh-2 & 13 24 23.98 & -42 54 10.7 & 13.56 & 19.87 & 1.270 & 582 & 81 + whh-3 & 13 24 32.17 & -43 10 56.9 & 14.10 & 19.59 & 1.470 & 636 & 78
+ whh-4 & 13 24 40.60 & -43 13 18.1 & 14.89 & 19.13 & 1.927 & 685 & 43 + whh-5 & 13 24 44.58 & -43 02 47.3 & 8.04 & 19.50 & 1.386 & 671 & 41 + whh-6 & 13 24 47.37 & -42 57 51.2 & 8.06 & 19.75 & 1.931 & 712 & 67 + whh-7 & 13 25 05.02 & -42 57 15.0 & 5.68 & 17.43 & 1.863 & 722 & 20 + whh-8 & 13 25 07.62 & -43 01 15.2 & 3.66 & 18.05 & 2.026 & 690 & 32 + whh-9 & 13 25 08.51 & -43 02 57.4 & 3.93 & 18.90 & 1.979 & 315 & 100
+ whh-10 & 13 25 12.84 & -42 56 59.8 & 4.95 & 19.70 & 1.334 & 664 & 141 + whh-11 & 13 25 14.24 & -43 07 23.5 & 6.71 & 19.55 & 2.066 & 562 & 55 + whh-12 & 13 25 18.27 & -42 53 04.8 & 8.25 & 19.31 & 1.521 & 558 & 99 + whh-13 & 13 25 21.29 & -42 49 17.7 & 11.91 & 19.34 & 1.854 & 467 & 47 + whh-14 & 13 25 25.49 & -42 56 31.2 & 4.64 & 20.28 & 1.300 & 436 & 80 + whh-15 & 13 25 26.78 & -42 52 39.9 & 8.48 & 20.13 & 1.153 & 506 & 57 + whh-16 & 13 25 27.97 & -43 04 02.2 & 2.89 & 19.20 & 2.036 & 749 & 80 + whh-17 & 13 25 29.25 & -42 57 47.1 & 3.38 & 18.73 & 1.409 & 588 & 127 + whh-18 & 13 25 30.07 & -42 56 46.9 & 4.39 & 18.48 & 1.752 & 771 & 31 + whh-19 & 13 25 31.03 & -42 50 14.9 & 10.92 & 17.51 & 1.177 & 470 & 49 + whh-20 & 13 25 34.36 & -42 51 05.9 & 10.12 & 19.19 & 1.517 & 242 & 54 + whh-21 & 13 25 35.22 & -43 12 01.5 & 10.97 & 19.22 & 0.994 & 243 & 61 + whh-22 & 13 25 35.31 & -43 05 29.0 & 4.56 & 18.63 & 1.631 & 492 & 126 + whh-23 & 13 25 45.90 & -42 57 20.2 & 5.07 & 18.60 & 1.137 & 286 & 63 + whh-24 & 13 25 46.00 & -42 56 53.0 & 5.43 & 19.85 & 1.288 & 566 & 48 + whh-25 & 13 25 50.34 & -43 04 08.2 & 5.12 & 20.17 & 1.621 & 551 & 95 + whh-26 & 13 25 56.59 & -42 51 46.6 & 10.76 & 19.16 & 1.762 & 412 & 36 + whh-27 & 13 26 12.82 & -43 09 09.2 & 11.51 & 18.73 & 1.885 & 545 & 60 + whh-28 & 13 26 14.18 & -43 08 30.4 & 11.25 & 19.32 & 1.414 & 506 & 123 + whh-29 & 13 26 22.08 & -43 09 10.7 & 12.79 & 19.98 & 1.538 & 505 & 78 + whh-30 & 13 26 23.60 & -43 03 43.9 & 10.55 & 19.56 & 1.488 & 470 & 66 + whh-31 & 13 26 41.43 & -43 11 25.0 & 16.96 & 19.65 & 1.618 & 573 & 67 + 36 & 13 24 02.55 & -42 49 29.4 & 64 & 42 & star + 3 & 13 24 28.96 & -43 10 58.3 & -24 & 23 & star + 8 & 13 24 52.11 & -43 08 24.3 & -1 & 26 & star + 19 & 13 24 56.46 & -43 06 34.7 & -46 & 24 & star + 11 & 13 25 04.20 & -43 02 29.3 & -30 & 26 & star + 14 & 13 25 08.59 & -43 07 43.4 & -17 & 47 & star + 13 & 13 25 10.06 & -43 07 40.7 & -30 & 21 & star + 26 & 13 25 17.79 & -43 13 05.6 & -70 & 42 & star + 25 & 13 25 24.37 & -43 12 23.1 & -13 & 23 & star + 62 & 13 25 25.10 & -43 12 27.0 & -82 & 79 & star + 27 & 13 25 42.62 & -43 10 21.4 & -20 & 32 & star + 77 & 13 25 47.98 & -43 00 57.6 & 106 & 106 & star + 4 & 13 25 55.77 & -43 03 00.9 & 18 & 21 & star + 1 & 13 25 56.40 & -42 59 37.3 & 40 & 23 & star + 2 & 13 25 57.38 & -42 59 48.5 & 1 & 22 & star + 6 & 13 25 58.39 & -42 55 27.6 & -22 & 21 & star + 23 & 13 25 58.44 & -43 03 06.4 & -3 & 25 & star + 20 & 13 26 10.41 & -43 00 53.6 & -11 & 26 & star + 73 & 13 26 15.41 & -42 57 46.8 & 3 & 118 & star + 56 & 13 26 17.81 & -43 00 29.7 & -60 & 37 & star + 17 & 13 26 20.74 & -43 09 13.2 & 51 & 33 & star + 89 & 13 26 21.11 & -43 03 08.2 & -85 & 99 & star + 61 & 13 26 32.79 & -43 02 59.4 & 31 & 50 & star + 38 & 13 26 33.08 & -43 03 14.0 & -91 & 40 & star + 283 & 13 24 04.50 & -42 48 16.2 & 47636 & 81 & galaxy + 285 & 13 24 05.49 & -42 47 24.1 & 4729 & 124 & galaxy + 290 & 13 24 13.26 & -42 57 42.1 & - & - & pc + 94 & 13 24 28.15 & -42 53 04.6 & 35382 & 178 & galaxy + 67 & 13 24 28.46 & -43 14 56.3 & - & - & pc + 122 & 13 24 29.70 & -43 02 06.4 & - & - & pc + 43/44 & 13 24 30.86 & -42 52 40.8 & 35527 & 70&galaxy + 117 & 13 24 34.38 & -42 57 32.4 & - & - & pc + 127 & 13 24 34.61 & -43 12 50.5 & 7717 & 194 & galaxy + 133 & 13 24 35.30 & -43 06 15.4 & - & - & pc + 119 & 13 24 48.71 & -42 52 35.5 & - & - & pc + 298 & 13 25 01.09 & -42 48 21.9 & - & - & pc + 136 & 13 25 01.16 & -43 11 59.0 & - & - & pc + 299 & 13 25 01.29 & -42 54 26.5 & 2544 & 92 & galaxy + 132 & 13 25 23.45 & -42 53 26.2 & 2017 & 78 & galaxy + 108 & 13 25 25.73 & -43 05 16.6 & - & - & pc + 103 & 13 25 28.72 & -42 50 12.3 & - & - & pc + 314 & 13 25 37.53 & -42 55 33.5 & - & - & pc + 116 & 13 25 39.63 & -43 04 01.4 & 32124 & 253 & galaxy + 123 & 13 25 41.37 & -43 12 12.6 & 31018 & 230 & galaxy + 317 & 13 25 45.63 & -43 01 15.5 & 5850 & 91 & galaxy + 72 & 13 25 49.23 & -43 00 02.2 & 7349 & 87 & galaxy + 90 & 13 25 49.26 & -43 02 20.4 & 9942 & 113 & galaxy + 135 & 13 25 49.95 & -43 02 26.3 & 38298 & 114 & galaxy + 107 & 13 25 56.63 & -42 58 45.5 & - & - & pc + 101 & 13 26 02.24 & -43 08 55.7 & - & - & pc + 109/110 & 13 26 02.23 & -43 17 34.8 & - & - & pc + 134 & 13 26 17.27 & -43 06 39.3 & 24742 & 79 & galaxy + 113 & 13 26 21.29 & -42 57 19.1 & - & - & pc + 125 & 13 26 23.64 & -43 00 45.6 & - & - & pc + 100 & 13 26 50.44 & -42 50 36.4 & - & - & pc + c2 & 13 24 51.47 & -43 12 11.2 & 12.86 & 18.38 & 1.546 & 606 & 38 + c3 & 13 24 58.21 & -42 56 10.0 & 7.33 & 17.68 & 1.940 & 559 & 20 + c4 & 13 25 01.81 & -43 09 25.5 & 9.53 & 17.88 & 1.451 & 687 & 49 + c6 & 13 25 22.19 & -43 02 45.6 & 1.89 & 16.83 & - & 846 & 18 + c7 & 13 26 05.40 & -42 56 32.4 & 8.30 & 16.90 & 1.533 & 599 & 15 + c10 & 13 24 48.04 & -43 08 14.3 & 10.13 & 18.30 & 1.727 & 856 & 33 + c11 & 13 24 54.70 & -43 01 21.6 & 6.02 & 17.67 & 2.011 & 765 & 15 + c13 & 13 25 06.22 & -43 15 11.6 & 14.58 & 18.48 & 1.703 & 599 & 25 + c14 & 13 25 10.49 & -42 44 52.8 & 16.56 & 17.75 & 1.655 & 691 & 19 + c15 & 13 25 30.39 & -43 11 49.6 & 10.69 & 18.43 & 1.881 & 672 & 33
+ c17 & 13 25 39.72 & -42 55 59.1 & 5.62 & 17.49 & 1.422 & 785 & 18 + c18 & 13 25 39.86 & -43 05 01.8 & 4.48 & 17.29 & 1.603 & 476 & 19 + c19 & 13 25 43.38 & -43 07 22.9 & 6.87 & 17.87 & 0.324 & 607 & 18 + c20 & 13 25 49.68 & -42 54 49.3 & 7.50 & 17.89 & 1.472 & 745 & 16 + c21 & 13 25 52.73 & -43 05 46.5 & 6.52 & 17.62 & 1.576 & 461 & 96 + c26 & 13 26 15.25 & -42 48 29.4 & 15.36 & 17.94 & 2.068 & 370 & 21 + c30 & 13 24 54.35 & -42 53 24.7 & 9.84 & 17.05 & 1.788 & 786 & 17 + c32 & 13 25 03.37 & -42 50 46.1 & 11.29 & 18.22 & 2.006 & 708 & 32 + c33 & 13 25 16.26 & -42 50 53.3 & 10.47 & 18.67 & - & 520 & 41 + c36 & 13 26 07.73 & -42 52 00.2 & 11.72 & 18.24 & 1.378 & 696 & 43 + c37 & 13 26 10.57 & -42 53 42.7 & 10.81 & 18.17 & 1.691 & 612 & 21 + c38 & 13 26 23.77 & -42 54 01.1 & 12.50 & 18.34 & 1.544 & 397 & 23 + c39 & 13 26 42.00 & -43 07 44.9 & 15.11 & 17.28 & 1.647 & 247 & 26 + c44 & 13 25 31.73 & -43 19 22.8 & 18.25 & 18.48 & 1.441 & 515 & 43 + c47 & 13 25 49.94 & -42 52 09.3 & 9.88 & 18.80 & 1.402 & 592 & 34 + c48 & 13 25 49.83 & -42 50 15.1 & 11.63 & 18.65 & 1.281 & 557 & 77 + c50 & 13 26 19.64 & -43 03 18.6 & 9.75 & 18.59 & 1.633 & 605 & 90
+ g271 & 13 25 13.94 & -42 57 42.5 & 4.25 & 18.80 & 1.092 & 263 & 98 + g342 & 13 25 05.75 & -42 59 00.3 & 4.53 & 18.20 & - & 565 & 37 + g369 & 13 24 57.49 & -42 59 23.3 & 5.78 & 18.80 & 1.436 & 527 & 58 + pff_gc-018 & 13 24 47.08 & -43 06 01.7 & 8.87 & 18.91 & 1.603 & 554 & 44 + pff_gc-041 & 13 25 11.14 & -43 03 09.6 & 3.62 & 19.61 & 1.471 & 432 & 179 + pff_gc-042 & 13 25 12.19 & -43 16 33.7 & 15.67 & 18.97 & 1.352 & 678 & 87 + pff_gc-056 & 13 25 32.80 & -42 56 24.3 & 4.84 & 18.64 & 1.286 & 275 & 44 + pff_gc-058 & 13 25 35.12 & -42 56 45.1 & 4.60 & 18.68 & 1.259 & 376 & 78 + pff_gc-060 & 13 25 42.42 & -42 59 02.5 & 3.43 & 18.93 & 1.634 & 818 & 96 + pff_gc-064 & 13 25 43.89 & -42 50 42.5 & 10.85 & 19.80 & 1.758 & 556 & 63 + pff_gc-073 & 13 25 52.76 & -42 58 41.6 & 5.21 & 20.10 & 1.759 & 470 & 189 + pff_gc-075 & 13 25 53.47 & -43 03 56.6 & 5.49 & 19.61 & 1.768 & 764 & 59 + pff_gc-076 & 13 25 53.74 & -43 19 48.5 & 19.26 & 19.07 & 1.923 & 351 & 39 + pff_gc-083 & 13 26 01.81 & -42 58 15.0 & 6.89 & 20.04 & 1.683 & 473 & 53 + pff_gc-085 & 13 26 06.39 & -43 00 38.1 & 7.11 & 19.63 & 1.262 & 469 & 108 + pff_gc-089 & 13 26 20.19 & -43 10 35.7 & 13.48 & 18.95 & 1.324 & 553 & 103 + | we present radial velocity measurements for 74 globular clusters ( gcs ) in the nearby giant elliptical ngc 5128 , of which 31 are newly discovered clusters .
all the gc candidates were taken from the list of possible new clusters given in the @xcite photometric survey .
in addition to the newly confirmed clusters , we identified 24 definite foreground stars and 31 probable background galaxies . from a combined list of 299
known gcs in ngc 5128 with measured radial velocities and metallicity - sensitive @xmath0 photometric indices , we construct a new metallicity distribution function ( mdf ) for the cluster system .
the mdf shows an approximately bimodal form , with centroids at @xmath1 } = -1.46 $ ] and @xmath2 , and with nearly equal numbers of metal - poor and metal - rich clusters in the two modes .
however , there are many intermediate - color clusters in the distribution , and the fainter clusters tend to have a higher proportion of red clusters .
these features of the mdf may indicate a widespread age range within the cluster system as well as an intrinsically broad metallicity spread . |
due to such unique properties as the high intensity , high degree of collimation , and wide spectral range ( see @xcite-@xcite and references therein ) synchrotron radiation ( sr ) serves as an extraordinary research tool for advanced studies in both the fundamental and applied sciences .
nowadays it is used worldwide by thousands of scientists in many disciplines like physics , chemistry , material science and structural biology .
these applications motivate the importance of analyzing various mechanisms for control of sr parameters . from this point of view
it is of interest to study the influence of medium on spectral and angular distributions of sr .
investigations of this kind are also important for a number of astrophysical problems @xcite .
the characteristics of high - energy electromagnetic processes in the presence of material are essentially changed by giving rise to new types of phenomena , the well - known examples of which are cherenkov @xcite-@xcite and diffraction radiation .
the operation of many devices intended for production of electromagnetic radiation is based on interactions of relativistic electrons with matter ( see , e.g. , @xcite ) . as was shown in @xcite , where sr from a charged particle rotating in a homogeneous medium was considered
( see also @xcite ) , the interference between sr and cherenkov radiation ( cr ) had interesting consequences .
new interesting phenomena occur in case of inhomogeneous media .
a well - known example here is the transition radiation . in particular ,
the interfaces between media can be used for monitoring the flows of radiation emitted in various systems . in a series of papers initiated in @xcite-@xcite
it was shown that the interference between sr and cr induced at boundaries of spherical or cylindrical configuration leads to interesting effects .
investigations of radiation from a charge rotating along an equatorial orbit about / inside a dielectric ball showed @xcite that when the cherenkov condition for the ball material and particle speed is satisfied , there appear high narrow peaks in the spectral distribution of the number of quanta emitted to outer space at some specific values of the ratio of ball - to - particle orbit radii . in the vicinity of these peaks
the radiated energy exceeds the corresponding value for the case of homogeneous and unbounded medium by several orders of magnitude .
however in @xcite the phenomena of absorption and dispersion of electromagnetic waves inside the ball material were not taken into account . the allowance for these phenomena
was made in @xcite . in the present paper
a new characteristic feature of such a high power radiation was revealed and a simple model that disclosed the physical cause of its generation was suggested .
it is worthwhile to point out that a similar phenomenon ( less pronounced ) takes place in the cylindrical symmetry case @xcite-@xcite .
e.g. , the radiation emitted ( i ) from a longitudinal charged oscillator moving with constant drift velocity along the cylinder axis , and ( ii ) from a charged particle moving along a circle around a dielectric cylinder or ( iii ) along a helical orbit inside the cylinder are investigated in @xcite-@xcite , @xcite and @xcite-@xcite respectively .
the latter type of motion is used in helical undulators for generating electromagnetic radiation in a narrow spectral interval .
it is shown that under similar cherenkov condition for permittivity of cylinder and the particle speed , high narrow peaks are present in the spectral - angular distribution for the number of radiated quanta . in the vicinity of these peaks
the radiated energy exceeds the corresponding value for homogeneous medium case by several orders of magnitude .
the content of paper is organized as follows : in section [ section2 ] the description of problem is given , and in section [ section3 ] the method of solution is described and the final analytical expression for the intensity of radiation from a relativistic particle rotating about a ball is given .
numerical results are presented in section [ section4 ] .
the characteristic features of phenomenon are discussed in section [ section5 ] .
the cause of radiation amplification is established in section section6 .
similar processes of amplification typical for a chain of equidistant particles / bunches , are considered in the next section . in the last section
the main results of paper are summarized .
now consider the uniform rotation of relativistic electron in the equatorial plane of a dielectric ball in the magnetic field in empty space ( see fig . [ fig1 ] ) .
the permittivity of medium is the following step function of radial coordinate r:@xmath0where @xmath1 is the radius of ball , and @xmath2 is the complex valued permittivity of the ball material .
we assume that the braking of electron due to the emission of radiation is compensated by an external influence ( e.g. , the electric force ) that compels the particle to turn uniformly in a circle .
it is convenient to introduce the following dimensionless quantity:@xmath3@xmath4(the number of emitted quanta ) .
here w@xmath5 is the energy radiated at frequency @xmath6 during one period @xmath7 of electron gyration , @xmath8 is the harmonic number , and @xmath9 is the energy of corresponding electromagnetic wave quantum .
it is known @xcite that if the space as a whole is filled with a transparent substance ( with constant @xmath10 ) then@xmath11@xmath12 , \label{nkinfinity}\]]where @xmath13 , @xmath14v@xmath15 and @xmath16 is the bessel function of integer order .
the case @xmath17 of this formula corresponds to the synchrotron radiation in vacuum ( see , e.g. , @xcite ) .
we aimed at : * calculation of @xmath18 , when @xmath19 is the step function ( [ epsr ] ) and * demonstration that in case of weak absorption of radiation in the ball material and at special choice of the values of @xmath1 and @xmath20 , the number of emitted quanta @xmath21
we have obtained @xcite ( see also @xcite ) the following expression @xmath22for the number of quanta emitted by the electron during one revolution . here
@xmath23 and @xmath24are dimensionless amplitudes describing the contributions of electrical ( _ e _ ) and magnetic ( _ h _ ) type multipoles respectively , @xmath25 are the associated legendre polynomials , and @xmath26 , @xmath27 are the following factors depending on @xmath28v@xmath29 and @xmath30 : @xmath31 , \quad u = k\text{v}/c,\quad u_{b}=k\text{v}\sqrt{\varepsilon _ { b}}/c , \notag \\ b_{l}\left ( e\right ) & = & \left ( l+1\right ) b_{l-1}\left ( h\right ) -lb_{l+1}\left ( h\right ) + x^{-2}\left ( 1-\varepsilon _ { b}\right ) \left [ j_{% \underline{l-1}}\left ( xu_{b}\right ) + j_{\underline{l+1}}\left ( xu_{b}\right ) \right ] \times \notag \\ & & \times \left [ h_{\underline{l-1}}\left ( u\right ) + h_{\underline{l+1}% } \left ( u\right ) \right ]
\frac{l\left ( l+1\right ) u_{b}j_{l}\left ( xu_{b}\right ) } { lz_{l-1}^{l}+\left ( l+1\right ) z_{l+1}^{l}}. \label{blh , ble}\end{aligned}\]]here @xmath32n@xmath33 , and @xmath34 , n@xmath33 are spherical bessel and neumann functions respectively . in ( [ blh ,
ble ] ) we used the following notations : @xmath35 in the absence of dielectric ball @xmath36 the calculations by means of our formula ( [ nkball ] ) give the same results as those obtained using the well - known synchrotron radiation theory formula ( [ nkinfinity ] ) .
now consider the radiation generated by the electron at some harmonic @xmath37 , e.g. , for @xmath38 . in fig.[fig2 ]
the number @xmath39 of emitted quanta was plotted versus the ratio @xmath40 of ball to electron orbit radii .
the energy and the orbit radius of the particle were taken to be @xmath41 and @xmath42 respectively . here
corresponding to the 8th harmonic is the radiation at frequency @xmath43 and wavelength @xmath44 ( in vacuum ) .
the number of quanta was calculated in @xcite by formula ( [ nkball ] ) with due regard for the dispersion and dielectric losses of energy inside the substance of ball under the assumption that the ball is made of melted quartz , the value of permittivity of which is @xcite : @xmath45 according to ( [ nkinfinity ] ) the number of quanta emitted on the 8th harmonic by electron with above values of @xmath46 and @xmath47 at rotation in a continuous , infinite and transparent medium with @xmath48would be @xcite @xmath49(the horizontal dashed line in fig.[fig2 ] ) . if the particle did not rotate , but moved rectilinearly in the same medium with energy @xmath41 , then during the period of time @xmath7 it would have emitted in a narrow frequency band @xmath50 @xmath51quanta ( see , e.g. , @xcite-@xcite ) .
it is evident that @xmath52 . in the absence of ball ( synchrotron radiation @xcite-@xcite ) @xmath53 according to fig.[fig2 ] , @xmath54 practically for all @xmath55 except for @xmath56 and @xmath57
. there are peaks in these ranges and for the highest one @xmath58 here more than 30-fold magnification of @xmath59 ( from @xmath60 to @xmath61 ) is due to the forced uniform rotation of electron around the ball under the action of an external energy source .
the corresponding value of ball radius @xmath62 and , consequently , the distance @xmath63 between the rotating electron and the ball surface should be @xmath64 .
apparently , @xmath65 similar results may be obtained for a series of other values of @xmath66 , as well as for electrons with @xmath67 energy and balls with @xmath68 and @xmath69 .
it is significant that the condition @xmath70holds \(a ) only for a number of values of @xmath71 and also \(b ) when simultaneously the cherenkov condition @xmath72for speed v@xmath73v@xmath74 of particle `` image '' on the ball surface ( see .
fig.[fig3 ] ) and the real part @xmath75 of the permittivity of ball material is satisfied . for all peaks ( [ nkball - nkinfinity ] ) the characteristic propagation lengths @xmath76 , after passage of which the amplitude of electromagnetic oscillations decreases in @xmath77 times , are always many times greater than the ball radius:@xmath78for this reason the third condition may be formulated as : \(c ) inside the ball the electromagnetic oscillations may pass the distances that exceed many times the radius @xmath1 of ball ( before oscillations are absorbed or leave the boundaries of ball ) .
if the electron is relativistic , then in addition to the synchrotron radiation it may also generate cherenkov radiation .
its rise is attributed to the fact that the field coupled with the relativistic electron partially penetrates the ball and rotates together with the particle . at small distances from the surface of ball : @xmath79 , the speed of this field displacement inside the ball may be larger than the phase speed @xmath80 of light in the ball material , and then cr with wavelengths @xcite-@xcite @xmath81is generated inside the ball . in this section for visual explanation of numerical results given in sections [ section4],[section5 ] we shall avail of the following simplified model .
let us assume that at the moment @xmath82 the electron is in point @xmath83 of orbit and its image is in point @xmath84 ( see .
fig.[fig3 ] ) . at this instant
the electron generates cr along with the synchrotron radiation , or to be more exact , generates cr pulse @xmath85 ( the wavy line ) in the vicinity of point @xmath84 inside the ball near the surface .
the direction of cr is characterized by some radiation angle @xmath86 .
we shall use the following ansatz that @xmath86is determined by the cherenkov condition@xmath87by analogy with the case of faster - than - light particle propagating rectilinearly in a continuous and infinite dielectric medium @xcite-10 . in ( [ costet ] ) v@xmath88v@xmath89 is the speed of particle `` image '' on the surface of ball . according to such a simplified approach
, the condition b of section [ section5 ] establishes that the electron generates cr inside the dielectric ball .
the pulse @xmath85 travels the distance @xmath90 to the vicinity of point @xmath91 by the surface of ball and is partially absorbed by the ball material . in ( [ costet ] )
@xmath92and , consequently , at least some part of radiation with the wavelength @xmath93(according to the laws of geometrical optics ) will be subjected to an `` almost total '' internal reflection at @xmath86 angle .
this process will continue till @xmath94 is totally absorbed and/or leaves the ball . at @xmath95 ( small adsorption ) the pulse @xmath96 `` will have time '' to reflect from the internal surface of ball @xmath97times and , consequently , to travel the distance @xmath98 inside the ball in agreement with condition c of the previous section . over the period of time @xmath99 , during which the oscillation amplitude in the pulse @xmath96 decreases in @xmath77 times
, the electron will make @xmath100 = \left [ \varepsilon _ { b}^{^{\prime } } /\pi k\varepsilon _ { b}^{^{\prime \prime } } \right ] > > 1 \label{number of revolution}\]]revolutions along the orbit ( @xmath101 $ ] means the integer part of number @xmath102 ) . in this process
the electron will appear in the vicinity of point @xmath83 many times at instants @xmath103 ; @xmath104 ; @xmath105 , and
its image will appear in the vicinity of point @xmath84 as much times again and each time will generate cr pulse .
all @xmath106 pulses @xmath107 with @xmath108 will propagate inside the confined space ( ball ) along identical `` trajectories '' . for the highest peak in fig.[fig2 ] @xmath109 and @xmath110 the aforesaid is applicable to each point along the electron orbit .
being superimposed in the confined space ( ball ) , the electromagnetic oscillations of numerous cr pulses will , naturally , damp each other . in this case the influence of ball will be small and the total radiation will not differ strongly from the synchrotron radiation .
this is precisely the case for the most values of @xmath55 ( see fig.[fig2 ] ) .
however , in some special cases the deviations from such a picture are possible . these take place in ranges @xmath111 and @xmath57 shown in fig.[fig2 ] . inside these ranges at some special choice of @xmath112(cf . with condition
a in section [ section5 ] ) the superposition of electromagnetic oscillations will be close to the in - phase one . in this case
the total field may prove considerably more strong than that at @xmath113 .
the force that brakes the gyration of electron will also increase and _ the extra work of external force compelling the electron to rotate uniformly along the orbit will be spent for generation of more powerful cr .
_ this is precisely the case at @xmath114 in fig.[fig2 ] . in section [ section4 ]
we determined the values of @xmath115 from the requirement of the presence of peak , @xmath116 , by numerical calculations by formula ( [ nkball ] ) , disregarding ( [ costet ] ) and neglecting the laws of geometrical optics . 1 .
thus , in general , the influence of ball on radiation from the particle is small and the total radiation differs slightly from synchrotron radiation .
2 . however , at some special values of the ratio @xmath117 high power cr will be generated .
3 . its rise is due to the fact that the electromagnetic oscillations of cr induced by the particle along all the trajectory are partially locked inside the ball and are superimposed in nondestructive way .
in 1996 and 1997 prof . wiedemann with co - workers experimentally observed a similar phenomenon for transition radiation ( tr ) from a chain of relativistic electron bunches of sub - picosecond duration from the stanford university short intense electron source ( sunshine ) . at the output of tr radiator ( metallic foil ) the radiation from one electron bunch was directed with the help of a special system of mirrors ( influence of boundaries ) back to the radiator so , that the radiation pulse arrived there at the instant of incidence of the other electron bunch to generate new tr pulse .
so , a superposition of electromagnetic pulses emitted at different moments of time took place within the formation zone of radiation . in the case of ball
, the superposition of electromagnetic pulses emitted at different moments of time also takes place in the radiation zone .
however , the pulses are emitted by the same particle ( not different particles ) .
[ section7.2 ] let us consider cr from a chain of equidistant charges moving along the axis of circular section waveguide filled with transparent and nondispersive dielectric ( e.g .. , teflon in the range of wavelengths @xmath118 26,27 ) . in fig.[fig4 ] the first charge
is shown to be in point @xmath83 .
let us trace the propagation of cr pulse @xmath119 induced by the first charge and emitted in small vicinity of point @xmath83 at the instant @xmath120 ( wave line in fig.[fig4 ] ) . at its propagation
the pulse @xmath121 at some instant @xmath122 will cross the trajectory of charges ( axis of waveguide ) in the vicinity of some point @xmath123 at the same angle @xmath86 , at which it was emitted . at that moment @xmath119
will fall behind its faster - than - light source that will be found more to the right at some point @xmath124 .
the second charge of the chain is more to the left at the distance of @xmath125 from the first one and if @xmath126 it will cross the vicinity of point @xmath91 simultaneously with the pulse @xmath127 .
cr pulse @xmath128 generated by the second charge in the vicinity of point @xmath91 will be formed and simultaneously interfere with @xmath119 for all @xmath129 .
if the direction of cr @xmath130v@xmath131 , and the radius @xmath132 of waveguide are known , one can easily calculate the distance @xmath133of a pulse @xmath119 from its source at the instant @xmath122 .
it is clear that at @xmath134 the generation of pulses in regions adjacent to the second , third and subsequent charges will take place simultaneously with their interference with the pulses emitted earlier by charges propagating ahead of them .
note , that a similar process may take place for all _ d _ multiple to the length of @xmath135 and , in particular , at @xmath136 .
the superposition of pulses may be accompanied by their suppression or amplification depending on this or that value of the phase difference .
for some values of @xmath137 the in - phase superimposition of oscillations is possible .
analytical and numerical calculations carried out in @xcite have shown that the in - phase superposition takes place if * @xmath138 ( the case of power quasi - monochromatic cr ) and * @xmath139 ( powerful cr in wider frequency band ) . in both the cases the pulses
are superimposed in - phase in regions directly adjacent to the radiating charges .
but for all that the superimposed pulses are generated by different charges . in case of ball the pulses are emitted by the same particle .
\1 ) we have studied the radiation from a relativistic electron at uniform rotation about a dielectric ball with due regard for dielectric losses of energy inside the ball material . here ,
in addition to the synchrotron radiation the electron may also emit cherenkov radiation .
its rise is attributed to the fact that the field coupled with the relativistic electron partially penetrates the ball and rotates together with the particle . at small distances from the surface of ball : @xmath79 , the speed of this field displacement may be larger than the phase speed of light in the ball material and then cherenkov radiation with wavelengths ( [ wavelength ] ) is emitted inside the ball . the peculiarities in total radiation at different harmonics @xmath140 due to the influence of matter and ball radius
are investigated theoretically .
\2 ) in general , the influence of ball on radiation from the particle is small and radiation differs slightly from synchrotron radiation . however , in case of weak absorption ( @xmath141 ) in the ball material at some harmonics with @xmath66 the electron may generate @xmath142 quanta of electromagnetic field during one rotation period
. e.g. , an electron of @xmath143 energy uniformly rotating at the distance of @xmath64 from the surface of fused quartz ball with radius @xmath62 will generate approximately one quantum @xmath144 of electromagnetic field at the wavelength @xmath145 ( 8th harmonic ) during one revolution ( see the highest peak in fig.fig2 ) .
this value is more than 30-fold greater than the similar value of @xmath146 for electron rotating in a continuous , infinite and transparent medium having the same real part @xmath147 of permittivity as that for the ball ( see ( [ nkbesk ] ) and the dashed line in fig.[fig2 ] ) .
\3 ) such a magnification of @xmath148 is due to the presence of external source of energy stimulating a uniform rotation of electron around the ball .
it is possible only if * the ratio between the radius of ball and that of electron orbit takes on a number of fixed values ( see , e.g. , ( [ n8ball / n8infinity ] ) ) * the cherenkov condition ( [ cherenkov ] ) for the speed of particle image on the ball surface and the ball material is satisfied and * inside the ball the cherenkov radiation field passes the distances that are many times greater than the radius @xmath1 of ball ( before the radiation will be absorbed or leave the ball boundaries ) .
\4 ) high power radiation rises due to the fact that electromagnetic oscillations of cherenkov radiation induced by the particle along all its trajectory are partially locked inside the ball and are superimposed in nondestructive way ( section [ section6 ] ) .
the authors are thankful to professor a.r .
mkrtchyan for general encouragement and a.a .
saharian , a.s .
kotanjyan for stimulating discussions .
one of the authors ( s.r.a . ) is especially thankful to s. bellucci for attention and assistance .
grigoryan , v national conference on application of the x - rays , synchrotron radiation , neutrons , and electrons for studying nano - materials ( rsne nano-2005 ) , 14 - 19 november 2005 , moscow , book of abstracts , p.302 ( in russian ) ; preprint hep - th/0512080 . | some characteristic features in the radiation from a relativistic electron uniformly rotating along an equatorial orbit around a dielectric ball have been studied .
it was shown that at some harmonics , in case of weak absorption of radiation in the ball material , the electron may generate radiation field quanta exceeding in several dozens of times those generated by electron rotating in a continuous , infinite and transparent medium having the same real part of permittivity as the ball material .
the rise of high power radiation is due to the fact that electromagnetic oscillations of cherenkov radiation induced along the trajectory of particle are partially locked inside the ball and superimposed in nondestructive way .
pacs : 41.60.bq ; 41.60ap keywords : relativistic electron ; cherenkov radiation ; synchrotron radiation |
the @xmath9 - @xmath10 mass difference @xmath11 is an interesting quantity to measure an explicit breaking of the chiral symmetry by the gauge coupling ( @xmath12 ) in the spontaneously broken phase of the chiral symmetry .
it in fact has been a prototype of the mass calculation of pseudo nambu - goldstone ( ng ) bosons in strong coupling gauge theories such as those in the technicolor theories @xcite and more recently in the little higgs models @xcite .
the sign as well as the absolute value of @xmath13 is an important issue ( `` vacuum alignment problem '' ) , since the negative sign would imply that the vacuum would align so as to break spontaneously the gauge symmetry ( @xmath12 in the case of pion ) which explicitly breaks the spontaneously broken symmetry .
such a situation never happens in the real - life qcd but may do in other theories .
hence this type of calculation plays a central role of the model buildings .
the first successful calculation of @xmath13 @xcite was done by the current algebra in conjunction with the weinberg spectral function sum rules @xcite saturated by the @xmath14 , @xmath7 and @xmath8 meson poles .
somewhat more elaborate calculation on this line was done combined with the qcd information @xcite .
recently , an effective field theory calculation without @xmath8 meson has been successfully done @xcite , based on the hidden local symmetry ( hls ) model @xcite at loop level @xcite and the wilsonian matching @xcite of the hls model with the qcd ( for reviews of the hls approach , see @xcite ) .
@xcite demonstrated that the hls model is a little higgs model with two sites and two links ( open moose ) whose locality of theory space forbids quadratic divergence in @xmath13 even without @xmath8 meson . on the other hand , direct qcd calculation of the @xmath13
has never been done except for the lattice simulation @xcite .
although the lattice qcd is a powerful method , it is still important to establish alternative qcd - based methods which are more intuitive and less dependent on the power of computers .
such methods will be crucial to the subjects which are difficult to be studied in the framework of the lattice simulation .
most relevant subjects are the dynamical origin of the electroweak symmetry breaking , qcd at finite density , etc . in this paper , we calculate in the chiral symmetric limit the @xmath13 simultaneously with the the pion decay constant @xmath15 and the qcd @xmath1 parameter @xcite ( or the gasser - leutwyler parameter @xmath2 @xcite ) within the framework of the the schwinger - dyson ( sd ) equation and the bethe - salpeter ( bs ) equation in the the improved ladder approximation .
we note that all these quantities are given by `` generalized '' weinberg spectral function sum rules for the difference between the vector current correlator @xmath3 and the axial - vector current correlator @xmath4 : @xmath16 is given by the weinberg first sum rule , the @xmath1 parameter by the das - mathur - okubo ( dmo ) sum rule @xcite or the `` zeroth weinberg sum rule '' , and @xmath13 by the das - guralnik - mathur - low - young ( dgmly ) sum rule @xcite or the `` third weinberg sum rule '' .
we then rewrite them in terms of the current correlators instead of the spectral functions and directly calculate @xmath17 on the same footing as @xmath15 and the qcd @xmath1 parameter through @xmath6 in the _ space - like momentum region_. such correlators are written in terms of the bs amplitudes which are given by solving the _ inhomogeneous _ bs ( ibs ) equation and the sd equation with the improved ladder approximation in the landau gauge both in the _ space - like region_. in contrast to our case , solving the bound state spectra by the homogenous bs equation in the time - like region is usually difficult , since it would need analytic continuation of the running coupling which is not an analytic function .
this kind of method has been extensively used to investigate masses and decay constants of low lying mesons @xcite in good agreement with the experiments . on the other hand ,
the same method failed to reproduce the qcd @xmath1 parameter consistently with the experiment @xcite .
however , we find that the qcd @xmath1 parameter is rather sensitive to the infrared ( ir ) cutoff parameter of the qcd running coupling .
we actually reproduce the qcd @xmath1 parameter consistently with the experiment by setting the ir cutoff parameter in the region never investigated in the previous works .
our parameter choice corresponds to taking the qcd scale parameter as @xmath18 , while it was chosen as @xmath19 in the previous results of the decay constants and the masses @xcite and of the qcd @xmath1 parameter @xcite , both values being considerably higher than the conventional @xmath20-@xmath21mev obtained in the @xmath22 scheme @xcite .
this actually corresponds to exploiting the `` scale ambiguity '' @xcite of @xmath23 in the sd and bs equations recently emphasized by hashimoto and tanabashi @xcite .
we then show that @xmath0 mass difference @xmath13 is calculated in rough agreement with the experimental value for the same parameter region as that reproducing the qcd @xmath1 parameter and @xmath15 . actually , although infrared dynamics is quite important for determination of @xmath15 and @xmath1 , it is not so important for determination of @xmath13 . in other words ,
@xmath15 and @xmath1 are sensitive to the infrared structure of the running coupling , while @xmath13 is not .
we also derive the masses and decay constants of @xmath7 meson and @xmath8 meson by fitting them to the calculated @xmath6 within our method using the pole saturated form of @xmath6 . these quantities
are also compared with experiments , which is another check of the validity of the calculation in the present analysis .
the calculation presented in this paper turns out to be the first example which derives @xmath13 directly from qcd without depending on the lattice simulation .
once we have checked the reliability of such a method in the ordinary qcd ( with three light flavors ) , we may apply it to the mass of pseudo ng bosons and vacuum alignment problem in other strong coupling gauge theories such as the large @xmath24 qcd , walking technicolor , little higgs , hot / dense qcd , etc .. this paper is organized as follows . in section [ sec : spectral ] we introduce spectral function sum rules : the dmo sum rule or the zeroth weinberg sum rule for @xmath1 parameter , the first weinberg sum rule for @xmath16 , the second weinberg sum rule , and the dgmly sum rule or the third weinberg sum rule for @xmath13 .
we then rewrite them in terms of the current correlators @xmath6 . in section [ sec : correlator - bs ]
we show how the current correlators are obtained from the bs amplitude which is calculated from the ibs equation in the space - like region . in section [ sec : ibs ] we introduce the ibs equation and show how to solve it numerically .
section [ sec : results ] is the main part of this paper : we first show the result of the ibs equation for qcd , and discuss the dependence of @xmath15 , @xmath1 and @xmath13 on the infrared structure of the running coupling .
then we compare them with the experimental values .
we also show results for masses and decay constants of @xmath7 and @xmath8 mesons . in section [ sec : summary ]
we give summary and discussions on applications to the large @xmath24 qcd and the electroweak symmetry breaking .
in this section we introduce the spectral function sum rules , which express the pion decay constant @xmath15 , the qcd @xmath1 parameter and @xmath13 in terms of @xmath6 in the space - like momentum region . here
we consider only the chiral symmetric limit of massless three flavors ( @xmath25 ) , the corrections of the finite quark mass being expected to be small for the quantities we consider .
let us begin with introducing the vector and axial - vector currents as @xmath26 @xmath27 where @xmath28 is the generator of @xmath29 normalized as @xmath30 and we consider @xmath31 . in the chiral symmetric limit the current correlator @xmath32
is given as @xmath33 the umezawa - kamefuchi - kllen - lehmann spectral representation for the current correlators are expressed as @xmath34 here , we decomposed the spectral function into the spin - one part @xmath35 and spin - zero part @xmath36 .
since the axial - vector current @xmath37 couples to the massless ng boson , the pion , while no massless particles couple to the vector current @xmath38 , we have @xmath39 where @xmath15 is the decay constant of the ng boson @xmath14 defined by @xmath40 using the spectral functions , we can write down the following `` generalized '' weinberg sum rules : @xmath41 & = & \frac{s}{4\pi}\ , , \label{eq : zeroth}\\ \int_{0}^{\infty } \frac{ds}{s}\ \left [ \rho_{v}^{(1)}(s ) - \rho_{a}^{(1)}(s ) \right ] & = & f_\pi^2\ , , \label{eq : first}\\ \int_{0}^{\infty } ds\ \left [ \rho_{v}^{(1)}(s ) - \rho_{a}^{(1)}(s ) \right ] & = & 0\ , , \label{eq : second}\\ - \frac{3 \alpha_{em}}{4\pi f_\pi^2 } \int_{0}^{\infty } ds \log{s}\ \left [ \rho_{v}^{(1)}(s ) - \rho_{a}^{(1)}(s ) \right ] & = & \delta m_\pi^2\ , , \label{eq : third}\end{aligned}\ ] ] where , @xmath1 is the qcd @xmath1 parameter @xcite , which in the case of @xmath42 is related to the gasser - leutwyler parameter @xmath2 @xcite as @xmath43 , @xmath44 is the coupling constant of the electromagnetic interaction , and @xmath13 is the @xmath45 mass difference defined by @xmath46 .
equations ( [ eq : first ] ) and ( [ eq : second ] ) are the first and second weinberg sum rules @xcite , respectively , and eqs .
( [ eq : zeroth ] ) is the dmo sum rule @xcite or often called the `` zeroth weinberg sum rule '' .
equation ( [ eq : third ] ) was derived by das et al .
@xcite and may be called the dgmly sum rule or the `` third weinberg sum rule '' . of course these relations in eqs .
( [ eq : zeroth ] ) , ( [ eq : first ] ) , ( [ eq : second ] ) , and ( [ eq : third ] ) make sense only if the integrals converge .
for example , convergence of the second sum rule ( [ eq : second ] ) requires that the difference between @xmath3 and @xmath4 must satisfy @xmath47\ \
\stackrel { ( q^2 \rightarrow \infty ) } { \longrightarrow}\ \ 0\ , , \label{eq : convergence}\ ] ] where @xmath48 is related to the space - like momentum as @xmath49 .
the asymptotic behavior of the current correlators can be calculated by the operator product expansion ( ope ) technique . in the chiral limit
, the form of @xmath50 in the ultraviolet region is estimated as follows @xcite : @xmath51 up to logarithm .
this means that @xmath50 indeed satisfies the condition ( [ eq : convergence ] ) and the relation in eq .
( [ eq : second ] ) makes sense . when the integral in the second sum rule converges , those in the first and zeroth sum rules are also convergent .
now we rewrite the zeroth and the first sum rules in terms of the current correlators as @xmath52 \right\vert_{q^2 = 0}\ , , \label{eq : s_parameter}\\ f_\pi^2 & = & \pi_{vv}(0 ) - \pi_{aa}(0)\ , , \label{eq : fpi}\end{aligned}\ ] ] and also the third sum rule : @xcite @xmath53\ , .
\label{eq : mass_diff}\ ] ] from eq.([eq : mass_diff ] ) , we can see that the asymptotic behavior ( [ eq : ope ] ) also guarantees convergence of the third sum rule . in ref .
@xcite the pion decay constant @xmath15 and the qcd @xmath1 parameter were calculated by the use of the above formulas .
however , @xmath13 was not calculated so far .
so the calculation presented in this paper turns out to be the first example which derives @xmath13 directly from qcd without depending on the lattice simulation .
moreover , there is another interest regarding @xmath13 related to the structure of the qcd vacuum , as we stressed in the introduction .
when we switch off the electromagnetic interaction and set masses of quarks to be zero , pions are identified with the exact ng bosons associated with the spontaneous breaking of @xmath54 chiral symmetry down to @xmath55 symmetry .
the existence of the electromagnetic interaction explicitly breaks the @xmath54 chiral symmetry , which makes @xmath9 and @xmath56 be the pseudo ng bosons . on the other hand ,
@xmath10 remains massless since the photon does not interact with @xmath10 .
the interesting point here is whether @xmath46 ( @xmath57 in the chiral limit ) becomes positive or negative , which is called `` vacuum alignment problem '' .
negative @xmath13 means that fluctuation of @xmath9 field around @xmath58 is unstable and the vacuum with @xmath58 is not a true vacuum .
if this is the case , @xmath9 has non - zero vacuum expectation value and @xmath12 symmetry is broken .
we know that , in the real world , @xmath13 is positive and the vacuum with @xmath58 is the true vacuum .
@xmath12 symmetry is not broken in the real world .
this is quite nontrivial fact resulting from the nonperturbative dynamics of the strong interaction .
so it is interesting to investigate whether we can reproduce positive @xmath13 which is realized in the real - life qcd .
in the previous section we have written down the qcd @xmath1 parameter , @xmath15 , and @xmath13 in terms of the current correlators . then , in this section , we show how the current correlators are obtained from the bs amplitude which will be calculated from the ibs equation . to derive properties of hadrons as boundstates , it is straightforward to perform calculations directly in the time - like momentum region . however , it is difficult to solve the bs equation and the sd equation in the time - like region since we have to carry out the analytic continuation of the running coupling from the space - like region to the time - like region . in the case of qcd ,
the one - loop running coupling is not an analytic function , so that we have to approximate it by some analytic function in order to perform the calculations in the time - like region .
however , here we need the bs amplitude only for the space - like region in order to calculate the current correlators for @xmath13 , @xmath15 and the qcd @xmath1 parameter . the bs amplitude @xmath59 ( @xmath60 , @xmath61 ) is defined in terms of the three - point vertex function as follows : @xmath62 where @xmath63 is the total momentum of the fermion and anti - fermion and @xmath64 is the relative one .
@xmath65 is the polarization vector defined by @xmath66 , @xmath67 , and @xmath68 are flavor , color and spinor indices , respectively . closing the fermion legs of the above three - point vertex function and taking the limit @xmath69
, we can express the current correlator in terms of the bs amplitude as follows : @xmath70 \label{eq : pi_jj}\ ] ] where @xmath71 and @xmath72 is the number of colors . in the above expression we averaged over the polarizations so that @xmath73 does not depend on the polarization .
we expand the bs amplitude @xmath74 in terms of the bispinor bases @xmath75 and the invariant amplitudes @xmath76 as @xmath77_{\alpha \beta } = \
\sum_{i=1}^{8 } \left [ \gamma_i^{(j)}(p;\hat{q},\epsilon ) \right]_{\alpha \beta } \chi^{(j)}_{i } ( p;q ) , \label{eq : chi - expanded}\ ] ] where @xmath78 . the bispinor bases can be chosen in a way that they have the same properties of spin , parity and charge conjugation as the corresponding current @xmath79 has .
we adopt the following bispinor bases for the vector vertex : @xmath80 ( p \cdot \hat q ) , \ \ \gamma^{(v)}_3 = \frac{1}{2 } [ \fsl{\epsilon},\fsl{\hat q } ] , \
\ \gamma^{(v)}_4 = \frac{1}{3!}[\fsl{\epsilon},\fsl{p},\fsl{\hat q } ] \\ & & \gamma^{(v)}_5 = ( \epsilon \cdot p ) , \ \
\gamma^{(v)}_6 = \fsl{p } ( \epsilon \cdot p ) , \ \
\gamma^{(v)}_7 = \fsl{\hat q}(p \cdot \hat q ) ( \epsilon \cdot p ) , \ \ \gamma^{(v)}_8 = \frac{1}{2 } [ \fsl{p},\fsl{\hat q } ] ( \epsilon \cdot p ) , \nonumber\end{aligned}\ ] ] where @xmath81 \equiv a[b , c ] + b[c , a ] + c[a , b]$ ] . for the axial - vector vertex we use @xmath82 \gamma_5 , \ \
\ \gamma^{(a)}_3 = \frac{1}{2 } [ \fsl{\epsilon},\fsl{\hat q}]\ ( p \cdot \hat q ) \ \gamma_5 , \nonumber\\ \gamma^{(a)}_4 & = & \frac{1}{3!}[\fsl{\epsilon},\fsl{p } , \fsl{\hat q } ] \
\gamma_5 , \ \
\ \gamma^{(a)}_5 = ( \epsilon \cdot p)\ ( p \cdot \hat q)\ \gamma_5 , \ \ \ \gamma^{(a)}_6 = \fsl{p } ( \epsilon \cdot p)\ \gamma_5 , \nonumber\\ \gamma^{(a)}_7 & = & \fsl{\hat q}\ ( \epsilon \cdot p)\ ( p \cdot \hat q ) \ \gamma_5 , \ \
\ \gamma^{(a)}_8 = \frac{1}{2 } [ \fsl{p},\fsl{\hat q } ] ( \epsilon \cdot p)\ ( p \cdot \hat q)\ \gamma_5.\end{aligned}\ ] ] from the above choice of the bases , we can easily show that all the invariant amplitudes @xmath83 are the even functions of @xmath84 using the charge conjugation property of the current . in the present analysis we fix the frame of reference in such a way that only the zero component of the total momentum @xmath63 becomes non - zero .
furthermore , we study the case where @xmath63 is in the space - like region .
then , it is convenient to parameterize the total momentum @xmath63 as @xmath85 for the relative momentum @xmath64 , we perform the wick rotation , and parameterize it by the real variables @xmath86 and @xmath87 as @xmath88 consequently , the invariant amplitudes @xmath76 become functions in @xmath86 and @xmath87 : @xmath89 from the charge conjugation properties for the bs amplitude @xmath59 and the bispinor bases defined above , the invariant amplitudes @xmath90 are shown to satisfy @xmath91 using this property of the invariant amplitudes , we rewrite eq .
( [ eq : pi_jj ] ) as @xmath92 , \label{eq : pivv}\ ] ] @xmath93
. \label{eq : piaa}\ ] ] here , we used the expanded form of the bs amplitude shown in eq .
( [ eq : chi - expanded ] ) and carried out the three dimensional angle integration . from eqs .
( [ eq : pivv ] ) and ( [ eq : piaa ] ) , the quantity @xmath6 is expressed as @xmath94,\nonumber\\ & = & \frac{n_c}{\pi^3 } \int_0^\infty du \int_0^\infty dx\ x^2 \bigg
[ - \left ( \chi^{(v)}_1(u , x;q ) + \chi^{(a)}_1(u , x;q ) \right ) \nonumber\\ & & \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ + \frac{x^2}{3 } \left ( \chi^{(v)}_6(u , x;q ) + \chi^{(a)}_6(u , x;q ) \right ) \ \bigg].\end{aligned}\ ] ] we note that , although either @xmath3 or @xmath4 is logarithmically divergent quantity , the difference @xmath6 becomes finite due to the cancellation of the divergence ensured by the chiral symmetry .
in this section we introduce the inhomogeneous bethe - salpeter ( ibs ) equation from which we calculate the bs amplitude defined in the previous section .
we also show the numerical method for solving the ibs equation .
the ibs equation is the self - consistent equation for the bs amplitude @xmath59 , and it is expressed as ( see fig .
[ fig : ibseq ] for graphical expression ) @xmath95 the kinetic part @xmath96 is given by @xmath97 where @xmath98 is the full fermion propagator @xmath99 .
( note that wave function renormalization factor @xmath100 becomes unity when we use the landau gauge . )
the bs kernel @xmath101 in the improved ladder approximation is expressed as @xmath102 where @xmath103 is the running coupling of qcd whose explicit form will be shown later . in the above expressions we used the tensor product notation @xmath104 and
the inner product notation @xmath105 the mass function of the quark propagator is obtained from the sd equation : @xmath106 it should be stressed that we must use the same kernel @xmath107 as that used in the ibs equation for consistency with the chiral symmetry @xcite . in this subsection
we transform the ibs equation in eq .
( [ eq : ibseq ] ) into the form by which we can solve it numerically .
first , we introduce the conjugate bispinor bases defined by @xmath108 multiplying these conjugate bispinor bases from left , taking the trace of spinor indices and summing over the polarizations , we rewrite eq .
( [ eq : ibseq ] ) into the following form : @xmath109 where the summation over the index @xmath110 is understood , and @xmath111 \ , \\ t^{(j)}_{ij}(u ,
x ) & = & \sum_{\epsilon } \mbox{tr } \left [ \bar\gamma_i^{(j)}(p;q,\epsilon ) t(p;q ) \gamma_j^{(j ) } ( p;q,\epsilon ) \right ] , \\ k^{(j)}_{ij}(u , x;v , y ) & = & \int_{-1}^{1 } d\cos\theta \ \sum_{\epsilon } \mbox{tr } \left [ \bar\gamma_i^{(j)}(p;q,\epsilon ) k(p , k ) \gamma_j^{(j)}(k;q,\epsilon ) \right ] \ , \end{aligned}\ ] ] with the real variables @xmath112 and @xmath113 introduced as @xmath114 here @xmath115 is the angle between the spatial components of @xmath116 and @xmath117 . using the property of @xmath83 in eq .
( [ eq : even - property ] ) , we restrict the integration range as @xmath118 : @xmath119 \chi_j^{(j)}(v , y).\ ] ] then , in the following , we treat all the variables @xmath86 , @xmath87 , @xmath112 and @xmath113 as positive values . to discretize the variables @xmath86 , @xmath87 , @xmath112 and @xmath113 we introduce new variables @xmath120 , @xmath121 , @xmath122 and @xmath123 as @xmath124 and set ultraviolet ( uv ) and infrared ( ir ) cutoffs as @xmath125 , \ \ x , y \ \in \ [ \lambda_x,\lambda_x ] .\ ] ] we discretize the variables @xmath120 and @xmath122 into @xmath126 points evenly , and @xmath121 and @xmath123 into @xmath127 points . then , the original variables are labeled as @xmath128 } = \exp\left[\lambda_u + d_u i_u \right ] , \
\ x_{[i_x ] } = \exp\left[\lambda_x + d_x i_x \right ] , \nonumber\\ & & v_{[i_v ] } = \exp\left[\lambda_u + d_u i_v \right ] , \ \ y_{[i_y ] } = \exp\left[\lambda_x + d_x i_y \right ] , \nonumber\end{aligned}\ ] ] where @xmath129 and @xmath130 .
the measures @xmath131 and @xmath132 are defined as @xmath133 as a result , the integration is converted into the summation : @xmath134 in order to avoid integrable singularities in the kernel @xmath135 at @xmath136 , we adopt the following four - splitting prescription @xcite : @xmath137 , \end{aligned}\ ] ] where @xmath138 , \ \
y_\pm = \exp\left[y \pm \frac{d_x}{4}\right ] .\ ] ] now that all the variables have become discrete and the original integral equation ( [ eq : ibseq ] ) has turned into a linear algebraic one , we are able to deal it numerically .
in this section , we show the results of calculations and give some discussions .
first , we discuss the dependence of the results on the infrared structure of the running coupling .
second , we compare the results with experimental values . for solving the ibs and sd equations , we have to fix the form of the running coupling @xmath139 which appears in the ibs and the sd equations .
we use the solution of the renormalization group equation for the qcd running coupling with one - loop approximation .
we regularize the infrared divergence of the one - loop running coupling by introducing the ir cutoff parameter @xmath140 as follows : @xmath141 \label{eq : running}\ ] ] where @xmath142 with @xmath24 being the number of flavors . as for the argument of the running coupling @xmath139
, we used the angle averaged form @xmath143 so that we can analytically carry out the angle integration in the sd and the ibs equations . in fig .
[ fig : mass_function ] , we plot the solutions of the sd equation for several values of @xmath144 .
.,height=283 ] for smaller value of @xmath144 , the value of mass functions @xmath145 in the infrared energy region becomes larger .
this is natural because smaller value of @xmath144 means larger running coupling in the infrared region .
when we solve the ibs equation , we use the following parameters : @xmath146 = \left [ -5.5 , 2.5 \right],\ ] ] @xmath147 = \left [ -2.5 , 2.5 \right],\ ] ] @xmath148 these parameters are chosen so that the dominant supports always lie within the energy region between uv and ir cutoffs for all values of @xmath144 which we use in the present analysis ( @xmath149-@xmath150 ) . in fig .
[ fig : pi ] , we plot the resultant values of @xmath50 for several values of @xmath144 . for several values of @xmath144 .
horizontal axis shows the square of the euclidean momentum normalized by the scale @xmath151 .
@xmath50 in the vertical axis is also normalized by @xmath151,height=377 ] horizontal axis shows the square of the euclidean momentum @xmath152 . in this figure ,
@xmath50 and @xmath48 are normalized by @xmath151 , and plotted as the dimensionless quantities . as we mentioned before
, either @xmath153 or @xmath154 is logarithmically divergent .
these divergences are expected to be canceled in @xmath155 ensured by the chiral symmetry .
figure [ fig : pi ] shows this cancellation actually occurs in the present calculation . would become @xmath156 .
figure [ fig : pi ] shows that @xmath157 is about @xmath158 at biggest , so that the divergences are actually canceled with each other . ] from the results shown in fig .
[ fig : pi ] , we can see that @xmath50 is dependent on the value of @xmath144 .
in other words , @xmath50 is sensitive to the infrared structure of the qcd running coupling .
once we obtain @xmath50 , we can calculate the qcd @xmath1 parameter , @xmath15 and @xmath13 by using eqs .
( [ eq : s_parameter ] ) , ( [ eq : fpi ] ) and ( [ eq : mass_diff ] ) . in table
[ tab : dimensionless ] , we list the resultant values of these quantities for several values of @xmath144 . here , we normalized dimensionfull quantities by @xmath159 , and expressed them as dimensionless quantities .
.resultant values of the qcd @xmath1 parameter , @xmath15 and @xmath160 for several values of @xmath144 .
dimensionfull quantities are normalized by @xmath159 and expressed as dimensionless quantities . [ cols="^,^,^,^,^,^,^",options="header " , ] resultant best fitted values
are listed in table [ tab : results-2 ] together with the experimental values . from the results shown in this table
, we can see that the masses and the decay constants of @xmath7 and @xmath8 mesons are not so much dependent on the infrared cutoff parameter @xmath144 compared with @xmath144 dependence of @xmath13 and @xmath1 .
when we choose @xmath149 , for which the values of @xmath13 and @xmath1 are predicted in good agreement with experiments , we obtained @xmath161 these are also in good agreement with experiments .
this confirms the reliability of calculations in the present analysis based on the improved ladder bs equation with small infrared cutoff parameter ( @xmath162 ) .
we have calculated the @xmath45 mass difference @xmath13 on the same footing as the pion decay constant @xmath15 and the qcd @xmath1 parameter through the vector and the axial - vector current correlators . for the calculations we solved inhomogeneous bs equations for the vector and axial - vector vertex functions together with sd equation for the quark mass function within the improved ladder approximation in the landau gauge .
we also obtained the masses and the decay constants of @xmath7 and @xmath8 mesons by fitting them to @xmath50 using the pole saturated form .
we showed that all of these quantities are simultaneously fit in agreement with the experiments when we take the ir cutoff parameter @xmath163 , which corresponds to taking @xmath164 this parameter choice of @xmath144 is fairly smaller than the ones investigated in the previous works @xcite , which corresponds to @xmath165 either value of @xmath23 above is substantially larger than that in the @xmath22 scheme @xcite : @xmath166 our results imply that the running coupling of qcd in the ir region should be taken larger than that considered so far , when we calculate not only @xmath15 but also the qcd @xmath1 parameter and @xmath13 in the framework of the bs and the sd equations in the improved ladder approximation .
apparently , choosing the ir cutoff parameter corresponds to exploiting the `` scale ambiguity '' @xcite of @xmath23 in the sd and bs equations as was recently emphasized in ref .
@xcite which proposed an interesting method to resolve the scale ambiguity in the the calculation of @xmath15 . in order to establish the approach of the sd and the bs equations in the improved ladder approximation , it is certainly desirable to apply the method of ref .
@xcite to our case , namely the calculations of @xmath13 and qcd @xmath1 parameter as well as @xmath15 . this will be done in future work .
the success of the analysis based on the bs and the sd equations in the real - life qcd in this paper well motivate us to apply this method for studies of other strong coupling gauge theories .
one of the most interesting examples of such is the large @xmath24 qcd ( see , e.g. , refs .
@xcite ) , which is a qcd with large number of massless flavors ( but not too large number as to destroy the asymptotic freedom ) .
it is expected that the chiral symmetry gets restored for a certain large number of massless flavors , which was in fact confirmed by the lattice simulations @xcite .
it was argued @xcite based on the effective field theory of hls @xcite that this chiral restoration of the large @xmath24 qcd is accompanied by the massless vector meson degenerate with the ng boson ( pseudoscalar meson ) as the chiral partners ( `` vector manifestation '' of chiral symmetry ) .
it was further argued @xcite that the hls model even in the real - life qcd behaves as a little higgs model @xcite with two sites and two links , whose locality of theory space forbids the quadratic divergence in @xmath13 which is nothing but the @xmath167 of the higgs in the framework of the little higgs models . in the large @xmath24 qcd @xmath13
becomes small compared with @xmath16 near the chiral restoration point , which may suggest that the large @xmath24 qcd may be considered as a uv completion of the little higgs model
. on the other hand , we found in the previous paper @xcite , by the explicit calculation based on the homogeneous bs and the sd equations in the improved ladder approximation , that masses of the scalar , vector , and axial - vector mesons in the large @xmath24 qcd are proportional to @xmath15 and vanish at the chiral restoration point , which implies somewhat different manifestation than the vector manifestation in the hls model .
( this behavior was also conjectured in ref . @xcite . )
thus it would be very useful to clarify the situation to calculate the @xmath1 parameter and @xmath13 in large @xmath24 qcd which have not yet been calculated directly from qcd . in the forthcoming paper
we shall calculate @xmath13 and @xmath1 parameter in large @xmath24 qcd by the same method presented in this paper .
we would like to thank yoshio kikukawa and masaharu tanabashi for discussions .
m.k . would like to thank thomas appelquist for warm hospitality and valuable discussion during his stay at yale university .
this work was supported in part by the jsps grant - in - aid for the scientific research ( b)(2 ) 14340072 ( k.y . and m.h . ) and by the 21st century coe program of nagoya university provided by jsps ( 15coeg01 ) .
n. arkani - hamed , a. g. cohen and h. georgi , phys .
b * 513 * , 232 ( 2001 ) n. arkani - hamed , a. g. cohen , e. katz , a. e. nelson , t. gregoire and j. g. wacker , jhep * 0208 * , 021 ( 2002 ) n. arkani - hamed , a. g. cohen , e. katz and a. e. nelson , jhep * 0207 * , 034 ( 2002 ) t. das , g. s. guralnik , v. s. mathur , f. e. low and j. e. young , phys .
lett . * 18 * , 759 ( 1967 ) s. weinberg , phys .
lett . * 18 * , 507 ( 1967 ) .
w. a. bardeen , j. bijnens and j. m. gerard , phys .
lett . * 62 * , 1343 ( 1989 ) .
m. harada , m. tanabashi and k. yamawaki , phys .
b * 568 * , 103 ( 2003 ) m.
bando , t. kugo , s. uehara , k. yamawaki and t. yanagida , phys .
lett . * 54 * , 1215 ( 1985 ) ; m. bando , t. kugo and k. yamawaki , nucl .
b * 259 * , 493 ( 1985 ) .
m. bando , t. kugo and k. yamawaki , prog .
phys . * 73 * , 1541 ( 1985 ) .
m. harada and k. yamawaki , phys .
b * 297 * , 151 ( 1992 ) m. tanabashi , phys .
b * 316 * , 534 ( 1993 ) m. harada and k. yamawaki , phys . rev .
d * 64 * , 014023 ( 2001 ) m.
bando , t. kugo and k. yamawaki , phys .
rept . * 164 * , 217 ( 1988 ) .
m. harada and k. yamawaki , phys .
rept . *
381 * , 1 ( 2003 ) a. duncan , e. eichten and h. thacker , phys .
lett . * 76 * , 3894 ( 1996 ) a. duncan , e. eichten and h. thacker , nucl .
suppl . * 53 * , 295 ( 1997 ) m. e. peskin and t. takeuchi , phys .
* 65 * , 964 ( 1990 ) ; phys .
d * 46 * , 381 ( 1992 ) .
j. gasser and h. leutwyler , annals phys . * 158 * , 142 ( 1984 ) .
nucl . phys .
b * 250 * , 465 ( 1985 ) .
t. das , v. s. mathur and s. okubo , phys .
* 19 * , 859 ( 1967 ) .
m. harada , m. kurachi and k. yamawaki , phys .
d * 68 * , 076001 ( 2003 ) k. i. aoki , m. bando , t. kugo , m. g. mitchard and h. nakatani , prog .
phys . * 84 * , 683 ( 1990 ) .
k. i. aoki , t. kugo and m. g. mitchard , phys .
b * 266 * , 467 ( 1991 ) . c. d. roberts and s. m. schmidt , prog . part .
* 45 * , s1 ( 2000 ) m. harada and y. yoshida , phys
. rev .
d * 50 * , 6902 ( 1994 ) a. j. buras , arxiv : hep - ph/9806471 . s. j. brodsky , g. p. lepage and p. b. mackenzie , phys .
d * 28 * , 228 ( 1983 ) .
m. hashimoto and m. tanabashi , arxiv : hep - ph/0210115 ; m. hashimoto and m. tanabashi , in _ proceedings of the 2002 international workshop on strong coupling gauge theories and effective field theories ( scgt 02 ) _ , nagoya , japan , dec 10 - 13 , 2002 , ed .
m. harada , y. kikukawa , and k. yamawaki ( world scientific pub .
co. , singapore 2003 ) , p.96 - 102 . c. w. bernard , a. duncan , j. losecco and s. weinberg , phys .
d * 12 * , 792 ( 1975 ) .
m. a. shifman , a. i. vainshtein and v. i. zakharov , nucl .
b * 147 * , 385 ( 1979 ) ; m. a. shifman , a. i. vainshtein and v. i. zakharov , nucl .
b * 147 * , 448 ( 1979 ) .
k. yamawaki , phys .
b * 118 * , 145 ( 1982 ) .
t. maskawa and h. nakajima , prog .
phys . * 52 * , 1326 ( 1974 ) ; t. maskawa and h. nakajima , prog .
phys . * 54 * , 860 ( 1975 ) .
t. kugo and m. g. mitchard , phys .
b * 282 * , 162 ( 1992 ) .
t. kugo and m. g. mitchard , phys .
b * 286 * , 355 ( 1992 ) .
m. bando , m. harada and t. kugo , prog .
phys . * 91 * , 927 ( 1994 ) h. pagels and s. stokar , phys . rev .
d * 20 * , 2947 ( 1979 ) .
n. isgur , c. morningstar and c. reader , phys .
d * 39 * , 1357 ( 1989 ) .
t. banks and a. zaks , nucl .
b * 196 * , 189 ( 1982 ) .
t. appelquist , j. terning and l. c. wijewardhana , phys .
lett . *
77 * , 1214 ( 1996 ) t. appelquist , a. ratnaweera , j. terning and l. c. wijewardhana , phys .
d * 58 * , 105017 ( 1998 ) v. a. miransky and k. yamawaki , phys .
d * 55 * , 5051 ( 1997 ) [ erratum - ibid .
d * 56 * , 3768 ( 1997 ) ] j. b. kogut and d. k. sinclair , nucl .
b * 295 * , 465 ( 1988 ) ;
f. r. brown , h. chen , n. h. christ , z. dong , r. d. mawhinney , w. schaffer and a. vaccarino , phys .
d * 46 * , 5655 ( 1992 ) ; y. iwasaki , k. kanaya , s. sakai and t. yoshie , phys .
lett . * 69 * , 21 ( 1992 ) ; y. iwasaki , k. kanaya , s. kaya , s. sakai and t. yoshie , nucl .
suppl . * 53 * , 449 ( 1997 ) ; prog .
suppl . * 131 * , 415 ( 1998 ) .
r. oehme and w. zimmermann , phys .
d * 21 * , 471 ( 1980 ) .
m. velkovsky and e. shuryak , phys .
b * 437 * , 398 ( 1998 ) .
m. harada and k. yamawaki , phys .
* 86 * , 757 ( 2001 ) r. s. chivukula , phys .
d * 55 * , 5238 ( 1997 ) | in the framework of the schwinger - dyson equation and the bethe - salpeter equation in the improved ladder approximation , we calculate the @xmath0 mass difference on the same footing as the pion decay constant and the qcd @xmath1 parameter ( or @xmath2 ) through the difference between the vector current correlator @xmath3 and the axial - vector current correlator @xmath4 .
we find that all the results can be fit to the experimental values for rather large @xmath5 which reflects the `` scale ambiguity '' . by fitting to the calculated data using the pole saturated form of @xmath6
, we also derive masses and decay constants of @xmath7 meson and @xmath8 meson , which we found are consistent with the experiments rather insensitively to the `` scale ambiguity '' . |
pure states are an important resource in quantum computation and communication algorithms @xcite . while state purification is possible via _ cooling _ , this may be impractical for several reasons , including long relaxation times , degenerate ground states , and the presence of several dephasing mechanisms .
a different purification process is possible via _ continuous measurement _ when the ( available to the observer ) quantum state will purify continuously according to the detector measurement result . here , the speed at which one can purify the state is set by the detector measurement rate , and the final purity will depend on how close the detector is to a quantum - limited ( 100% efficient ) one . this may become an important tool since continuous measurements are also at the heart of various quantum control applications@xcite , including quantum state stabilization via quantum feedback @xcite , preparation of entangled states @xcite , and for continuous error corrections @xcite . in recent years several groups have suggested rapid purification protocols based on continuous measurement and hamiltonian feedback @xcite , which makes it possible to considerably speed - up purification . for a single qubit ,
the problem was first analyzed by jacobs @xcite , who recognized that the rate of average purification can be enhanced using unitary transformation at each measurement time step , so as to make the state always orthogonal ( in the bloch sphere sense ) to the detector s measurement basis .
this feedback algorithm ( which has been rigorously shown to be optimal @xcite , and rederived in ref .
@xcite ) produces a factor of 2 speed - up in the high - purity limit , when the state approaches the bloch sphere surface .
the speed - up is by comparison with a no - feedback measurement , which can be completely understood classically ( see below ) .
this speed - up is a quantum mechanical effect since it is only possible if the system can exist in superposition of the eigenstates of the measured observable .
jacobs protocol maximizes the average purification rate in the sense that it minimizes the time @xmath16 it takes the average purity @xmath17 to reach a certain level @xmath3 . a different , but equally well motivated ,
aim was suggested by wiseman and ralph @xcite , that of minimizing the mean time @xmath18 for the purity @xmath19 to attain a certain level .
it was argued ( and later rigorously confirmed @xcite ) that the optimal protocol in this case is to rotate the state so that at each measurement time step it is aligned with the detector ( i.e. , making the state density matrix diagonal in the basis of the detector observable ) .
this is the classical protocol referred to above , and in the absence of any other dynamics , requires no feedback to realize . in this paper
we study a recently suggested purification protocol based on monitoring of the qubit state via simultaneous continuous measurement of the three non - commuting qubit operators @xcite @xmath0 , @xmath1 , @xmath2 ( fig.[schematic ] ) .
( a related protocol based on simultaneous measurement of these observables was first considered in ref.@xcite . ) in contrast to the above purification protocols there is no need to perform hamiltonian feedback . in this sense
it is an `` open loop '' protocol , similar to the `` random unitary control '' purification protocols introduced in ref .
@xcite for spin-@xmath20 systems .
though operationally different , the two protocols can be shown to be equivalent in the special ( @xmath21 ) case , which is another consequence of the isotropy of the qubit evolution discussed below . considering the rate of average purification under ideal measurements ( with detector efficiency @xmath4 ) the purification speed - up is @xmath5 , ( this speed - up is implicit in the results of ref .
@xcite ) . here
the comparison is with a _
single _ measurement of the same strength with no feedback ( the speed - up of @xmath22 quoted in ref .
@xcite comes from keeping the total measurement strength the same in the two protocols ) .
however , we show here that by the alternate metric the mean time @xmath18 of first passage the speed - up is only @xmath7 . both of these purification speed - ups can be understood via the isotropy of the qubit evolution @xcite in the bloch space that allows one to represent the three detectors by another equivalent detector triad , @xmath23 , @xmath24 , @xmath25 , measuring at each time moment along the state and in the complementary directions .
we also study the purification dynamics for inefficient detectors , @xmath26 . in the asymptotic high - purity limit ( @xmath10 )
the increase @xmath27 of the mean time of first passage @xmath18 shows a _ scaling behavior _
( i.e. it depends only on the ratio @xmath12 ) as seen from numerical calculations .
the increase @xmath28 is linear for relatively small inefficiency but grows exponentially , reaching @xmath14 for relatively large inefficiency .
for a quantum - limited detector , in the limit of infinite detector bandwidth , the evolution of the state @xmath29 of a quantum system due to weak continuous measurements of a variable @xmath30 can be described by the stochastic master equation ( sme ) @xcite : @xmath31\ , \rho + \sqrt{\frac{\gamma_0}{2}}\ , dw(t)\ , { \cal h}[x]\ , \rho .
\label{sme}\ ] ] here @xmath32\rho \equiv a\rho a^{\dag } - \frac{1}{2}(a^{\dag } a \rho + \rho a^{\dag } a)$ ] , @xmath33\rho \equiv a \rho + \rho a^{\dag } - \mbox{tr}[(a^{\dag } + a)\rho ] \rho$ ] , and @xmath34 is the increment of a wiener noise process . the internal hamiltonian evolution of the system is either neglected ( in case of a large measurement strength @xmath35 ) , or eliminated ( see , e.g. ref .
@xcite ) . the _ measurement rate _
@xmath35 sets the rate at which information ( about the final state ) is extracted , and thus the rate at which the system is projected onto a single eigenstate of @xmath30 @xcite ( provided the spectrum of @xmath30 is non - degenerate ) .
we will specialize on the measurement of a qubit and choose first @xmath36 , although some results below are valid for a general @xmath30 .
the measurement record in a small time interval @xmath37 is given by @xmath38 where @xmath39 $ ] , and @xmath40 is the same realization of the wiener increment that appears in eq .
( [ sme ] ) , thus providing explicit dependence of the state evolution on the specific measurement result .
the evolution generated via eq .
( [ sme ] ) corresponds to infinitesimal measurement operators ( see , e.g. refs .
since the same observable @xmath30 is measured at each time step , these operators commute , allowing one to integrate eq .
( [ sme ] ) exactly .
thus , for a finite time interval @xmath41 the final state will depend only on the time - averaged measurement record : @xmath42 introducing the measurement operator @xcite @xmath43}\ ] ] one obtains , using the method of un - normalized density matrices ( see , e.g. refs .
@xcite ) : @xmath44 here the probability distribution for @xmath45 is given by @xmath46 $ ] . in the basis of the observable @xmath30 , @xmath47
, the above result can be written as a quantum bayesian filter @xcite : the update of the diagonal density matrix elements @xmath48 will look exactly as a classical bayesian update of a `` probability distribution '' : @xmath49 , with gaussian likelihoods @xmath50 ; this was termed in ref .
@xcite the _ quantum - classical correspondence _ principle .
the outcomes @xmath51 will be gaussian distributed around the eigenvalues of @xmath30 , @xmath52 : @xmath53 $ ] with variance @xmath54 , and expected outcome probability @xmath55 .
the update of the non - diagonal matrix elements in ( [ povm ] ) will be according to the rule : @xmath56 provided the measurement is preformed by a quantum - limited detector ; this implies that a pure state will remain pure @xcite . from the above it
is clear that the measurement time @xmath57 to distinguish approximately two eigenvalues , @xmath58 is set by @xmath59 . for a qubit this gives @xmath60 @xcite . for further use
we also write down the quantum state evolution in terms of the bloch vector components of the state , @xmath61 , @xmath62 , @xmath63 . from eq .
( [ sme ] ) for the measurement of @xmath2 it follows : @xmath64 in these equations we also included the effect of detector non - ideality ( inefficiency ) @xmath65 , leading to a pure dephasing : in the ensemble averaged equations ( since eqs.([bloch-1-det ] ) are in the it form , averaging means just to drop the noise term , and corresponds to ignoring the detector results ) , @xmath35 is the decoherence rate due to an ideal detector of measurement rate @xmath35 , and @xmath66 is the detector efficiency ( ideality ) , defined as the ratio of @xmath35 to the total decoherence @xmath67 . for a single detector
the simplest model to describe the extra decoherence @xmath68 is to consider a second independent detector `` in parallel ''
( i.e. , measuring the same @xmath2 variable ) by adding terms similar to eq .
( [ sme ] ) with @xmath35 replaced by @xmath68 , and subsequently averaging over that detector output .
thus , the density matrix available for an observer who takes into account only the results of the first detector will be described by eqs.([bloch-1-det ] ) .
we consider first purification protocols via single - detector measurement with an ideal ( quantum - limited ) detector .
for a single detector measurement without feedback , the state evolution in the detector basis is essentially classical , eq .
( [ povm ] ) .
that is , if the density matrix begins diagonal in the measurement basis ( as it will be if it is a completely mixed state ) , it remains so and there is no way to distinguish @xmath69 from a classical probability distribution @xcite . for continuous
measurement one can consider the purity or entropy of the monitored state , and in this particular case the von neumann entropy of the state coincides with the shannon entropy ( see , e.g. refs .
@xcite ) : @xmath70 = - \left ( \rho_{11 } \ln{\rho_{11 } } + \rho_{22 } \ln{\rho_{22 } } \right)$ ] . in what follows we will consider the so called linear entropy ( see , e.g. @xcite ) , @xmath71 which is a monotonic function of @xmath72 . here
@xmath73 is the purity expressed through the bloch components of the state .
the corresponding equation for the purity ( at @xmath4 ) follows from eq .
( [ bloch-1-det ] ) : @xmath74 dt + 2 \sqrt{2\gamma_0}\ , z ( 1-p ) \ , dw \label{ito - dp-1det } .\ ] ] since purity @xmath19 , like entropy , is invariant under unitary transformations , without loss of generality one can say that measurement `` parallel to '' ( in the same basis as ) the state corresponds to @xmath75 and @xmath76 ^ 2 dt + \sqrt{2\gamma_0}\ , z [ 1-z^2 ] \ , dw \label{ito - dp0 } .\ ] ] so that the average change of purity is @xmath77 .
we note that the same result holds for non - ideal measurement , @xmath78 , since the detector s non - ideality affects only the evolution of the non - diagonal density matrix elements ; this means that a non - ideal detector will purify the state if the measurement is along the state .
it is clear that @xmath35 plays the role of a maximal classical purification ( information acquisition ) rate that happen when @xmath79 , i.e. when the two outcomes are equally likely ( see , e.g. , ref .
@xcite ) . by approaching @xmath80 , @xmath81 since
`` little information '' remains to be extracted to clarify that collapse has happened @xcite . from the discussion above ,
it is intuitively clear how the jacobs enhanced purification protocol works . given the state @xmath29 , one should continuously adapt the measurement basis [ or , equivalently , rotate the state to some @xmath82 so that the detector would perform measurement in a complementary direction with respect to the eigenbasis of the rotated state ( i.e. , measuring _
perpendicular to the state _
, in the bloch picture ) . in the detector basis
the rotated state again possesses _ equally likely outcomes _ , with @xmath83 .
as expected classically @xcite , this procedure maximizes the average purification rate since @xmath84 under rotation , while @xmath85 in eq .
( [ ito - dp-1det ] ) . however , the possibility to perform coherent rotations of the density matrix is of course a quantum mechanical effect .
the procedure also makes the purity evolution deterministic @xcite , i.e. the noise term in eq .
( [ ito - dp-1det ] ) is zeroed so that @xmath86 and @xmath87 under this protocol . from this one
can evaluate the time @xmath88 when the average purity @xmath17 reaches a given level @xmath3 : @xmath89 it should be noted that an attempt to evaluate the analogous time @xmath90 in the case of the classical measurement along the state , using @xmath91 from eq .
( [ ito - dp0 ] ) , will lead to a wrong scaling of @xmath92 .
the reason is that @xmath93 in general , and the evolving purity distribution @xmath94 is generally different from @xmath95-function @xcite .
the correct evaluation of @xmath90 is to calculate the average purity @xmath96 by taking into account the exact solution , eq .
( [ povm ] ) , of the stochastic evolution equations ( [ sme ] ) . in the high - purity limit this leads to @xcite @xmath97
this is exactly twice as long as the time in eq .
( [ tau_perp ] ) , which establishes the speed - up of 2 for the jacobs protocol .
we note however , that , unlike the case of parallel measurement , the perpendicular measurement will not purify to a completely pure state if the measurement is inefficient , as will be explored in sec .
[ sec:3c ] .
the reason is that the excess back - action in an inefficient perpendicular measurements results in decay in the coherences of the state . instead of considering the time @xmath16
at which the average purity @xmath17 reaches a certain level @xmath3 , one can also consider the average time @xmath98 for a system to attain that purity level , @xmath99 @xcite .
it was noted in ref .
@xcite that in many circumstances the time @xmath98 is a more useful quantity .
the reason is that the time @xmath100 at which @xmath99 , has a well - behaved statistics @xcite , in contrast with @xmath101 which has extremely long tails at relatively small values of @xmath19 .
that is , the averaged @xmath17 is strongly influenced by the rare cases that are slow to purify . because of this there is a substantial disagreement between @xmath16 and @xmath98 for a qubit .
it was shown in ref .
@xcite , however , that good agreement is found between @xmath98 and @xmath102 , defined as the time required for @xmath103\rangle$ ] to reach the certain level @xmath104 .
this is because taking the logarithm de - emphasizes the tails , and indeed for a qubit @xmath105 has near - normal distribution @xcite .
therefore , we consider the stochastic equation for the logarithm of the linear entropy @xmath106 .
it follows from eq .
( [ ito - dp-1det ] ) that : @xmath107 ( here , without loss of generality we have put @xmath108 , since a single @xmath2-measurement keeps the state of the qubit in a fixed meridional plane ) . in the wiseman - ralph feedback protocol @xcite one keeps the monitored state along the detector @xmath109-axis ( so @xmath110 , while @xmath111 is maximized at @xmath112 ) .
thus , the single detector purification is maximized and in the high - purity limit , @xmath113 , we obtain @xmath114 . using this , the average time ( of first passage ) @xmath98
is evaluated as @xmath115 and in the high - purity limit , @xmath116 @xmath117 which is half the size of @xmath88 , and 4 times shorter than @xmath90 . in what follows
we will use these results to understand the purification via three simultaneous complementary measurements without feedback .
we consider continuous measurement of the qubit complementary observables @xmath0 , @xmath1 , and @xmath2 by three independent linear detectors ( see fig . [ schematic ] ) . in principle , given a @xmath2-detector , the measurement of , say , @xmath0 can be implemented operationally by performing fast unitary rotation , @xmath118 , of the state towards the @xmath119-axis `` at the beginning '' of the measurement interval @xmath120 , then a continuous measurement with the @xmath109-detector and finally , a backwards transformation , @xmath121 `` at the end '' , in any infinitesimal measurement time step @xmath120 , as discussed in ref .
we also mention that a simultaneous measurement of the three observables is possible to implement , at least in principle , in a quantum optics setup @xcite ( e.g. , using a single atom in an optical cavity field ) . in the case of detectors measuring the qubit in the bases of non - commuting observables , it is impossible to use the quantum bayes rule ( [ povm ] ) for finite times because the measurement back - actions , @xmath122 , @xmath123 , @xmath124 , do not commute with each other .
the measurement back - actions commute only for small measurement time intervals @xmath120 so that one should apply the povm update in its differential form eq .
( [ sme ] ) and then sum up the contributions to the qubit evolution .
for the case of mutually unbiased observables @xmath0 , @xmath1 , and @xmath2 it is convenient to use the bloch vector components , @xmath125 , k = x , y , z\}$ ] determined by the time - dependent qubit density matrix @xmath126 . from eq .
( [ bloch-1-det ] ) for the influence due to @xmath2-measurement , the influences due to @xmath0 ( @xmath1)-measurement can be obtained by cyclic replacements @xmath127 , bearing in mind that the variables @xmath119(@xmath128 ) should play the same role as @xmath109 in a @xmath2-measurement . in this way we obtain the following evolution equations in it form ( see ref .
@xcite ) that take into account the three measurement records @xmath129 as in eq . ( [ result ] ) , and correspondingly introduce three independent wiener processes , @xmath130 : @xmath131 here @xmath132 are the total decoherence rates for each detector , including individual dephasings @xmath133 and measurement rates @xmath134 .
note that a pure dephasing , say in the @xmath119-basis , causes the decay of the @xmath109- and @xmath128-components .
( [ 3det - itoy ] ) and ( [ 3det - itoz ] ) are simply obtained by cyclic permutation of the variables in eq .
( [ 3det - itox ] ) . in what follows
we consider the simplest ( but still rich ) case of three identical detectors : @xmath135 and @xmath136 .
then the qubit evolution ( [ 3det - itox])([3det - itoz ] ) can be rewritten in a vector form as @xmath137\right ] \right\ } \label{ito - vec2 } , \end{aligned}\ ] ] where @xmath138 is the vector of wiener increments corresponding to the vector of results @xmath139 , similar to eq .
( [ result ] ) .
the evolution ( [ ito - vec2 ] ) is invariant under arbitrary rotations ( see fig . [ fig2 ] ) of the coordinate system in bloch space @xcite ( as is the ensemble averaged evolution : @xmath140 ) .
hence while measurement of only a single observable @xmath141 `` attracts '' the qubit state to one of the corresponding eigenvectors , the simultaneous measurement of @xmath0 , @xmath1 , and @xmath2 leads to no preferable direction in the bloch space . in ref .
@xcite this isotropy of the qubit evolution was shown to lead to a locally isotropic brownian diffusion of the direction of the bloch vector .
in particular , for pure states under ideal measurements , the point on the bloch sphere diffuses isotropically @xcite with a diffusion coefficient @xmath35 @xcite .
the evolution isotropy was used then to construct simple quantum state estimations @xcite . in what follows
, we will use the qubit evolution isotropy to understand the qubit purification dynamics .
( [ ito - vec2 ] ) shows that the measurement term component along the radius is @xmath142 and vanishes on the sphere .
the decreasing @xmath143-dependent coefficient suggests that @xmath144 is an attractor of the random evolution .
thus , the qubit state will purify in the ideal case ( @xmath145 ) for any realization of the measurement process . to establish further the purification dynamics , we start with eq .
( [ ito - vec2 ] ) and transform it to polar coordinates , @xmath143 , @xmath146 , @xmath147 .
the equation for the radius decouples from the other two and reads @xcite : @xmath148 where @xmath149 is the projection of the noise vector onto the bloch vector direction , @xmath150 , and the normalization of the noise is @xmath151 . as seen from eq .
( [ ito - radial ] ) , if @xmath152 , then on average @xmath153 , which means state purification for ideal measurements , @xmath4 .
since the equation for @xmath143 is singular at the origin , it will be more convenient to consider further the dynamics of the purity @xmath154 . in this choice @xmath155
, i.e. , a totally mixed state corresponds to @xmath156 .
we find @xmath157 for an ideal measurement , @xmath4 , and starting from a non - pure initial state , the qubit will purify ( @xmath158 ) on a time scale of the order of @xmath159 . for a non - ideal measurement , @xmath78 , purity will continue to fluctuate around a stationary average value , @xmath160 . to quantify the purity dynamics and asymptotic purity distribution we first consider the fokker - planck equation ( fpe ) corresponding to eq .
( [ ito - dp-3det ] ) . using the standard coefficients
( see ref .
@xcite ) : @xmath161
\label{fp - a - coeff}\\ & & b(p ) = 8\gamma_0\ , ( 2 p - 1 ) ( 1-p)^2 \label{fp - b - coeff}\end{aligned}\ ] ] the fpe reads @xmath162 + \frac{1}{2}\,\frac{\partial^2}{\partial p^2}\ , \left [ b(p ) { \cal p}(p , t)\right],\ ] ] and the initial distribution is taken to be @xmath163 . at @xmath164 the purity reaches a stationary distribution @xcite @xmath165 } , \label{fp - pur - stationary}\ ] ] where @xmath166 is the normalization . for @xmath167 , the stationary distribution @xmath168 approaches the @xmath95-function at @xmath169 .
the purification dynamics can be approximated using the ensemble - averaged purification rate @xcite obtained from the above it equation ( [ ito - dp-3det ] ) for purity : @xmath170 now consider a naive approach to integrating eq .
( [ av - pur - rate ] ) , in which we replace @xmath19 by @xmath17 everywhere it appears on the right - hand - side : @xmath171 \ , dt \label{av - pur - rate - naive } .\ ] ] contrary to the single - detector measurement , it can be shown , by numerically solving the fpe , that the evolution of the average purity @xmath17 obtained from the naive eq .
( [ av - pur - rate - naive ] ) is very close to the true average @xmath172 , obtained from fpe .
( the reasons for this difference will be discussed below . ) numerically , the approximation is best in the high purity limit ( @xmath173 , @xmath174 ) and for almost ideal detectors , @xmath175 . in the high - purity limit
we can discard terms of order @xmath176 ( i.e. the final term ) so that eq .
( [ av - pur - rate - naive ] ) becomes a simple linear equation . for ideal measurements
the time when the average purity reaches the level of @xmath177 is thus @xmath178 which is four times shorter than the standard time @xmath90 of classical measurements , eq .
( [ tau_parallel ] ) .
this means with three orthogonal measurements we obtain 4 times speed - up as compared to 2 times speed - up of the jacobs ( single measurement ) purification protocol .
this result can be understood as follows . by the isotropy of the qubit evolution under three complementary measurements , in each time moment @xmath179
one can represent the measurements in the @xmath119 , @xmath128 , @xmath109 directions by measurements in an equivalent triple of directions @xmath180 , @xmath181 , @xmath182 ( in the bloch space , fig .
[ fig2 ] ) , chosen so that @xmath182 is parallel to the state ( i.e. , in the basis of @xmath25 the density matrix @xmath29 is diagonal ) , while directions @xmath180 , @xmath181 are perpendicular to the state ; of course these chosen directions must change in time according to the evolution of the state , @xmath29 .
the measurements in the @xmath180 , @xmath181 directions in the time interval @xmath37 are termed `` good '' measurements as they are precisely the unbiassed measurements of jacobs that maximize the average purification rate , and the @xmath182-measurement along the state would be termed `` bad '' by obvious reason .
this observation could be confirmed by rewriting eq .
( [ av - pur - rate ] ) for @xmath4 in the following way : @xmath183 where the first two terms correspond to the deterministic change , @xmath184 , eq .
( [ dp_deterministic ] ) , and the last term coincides with the average change of @xmath185 due to measurement along the state , eq .
( [ ito - dp0 ] ) . in the high purity limit
, the last term is suppressed and the two `` good '' measurements contribute each a speed - up of 2 that amount to a total speed - up @xcite of 4 .
moreover , this makes clear why the naive approach ( replacing @xmath186 by @xmath187 ) to calculating the mean purity works for the isotropic measurement : jacob s protocol gives deterministic growth of the purity , so that @xmath188 . for isotropic measurement the purification in the mean is dominated by the perpendicular measurements , as just shown .
therefore the purification is approximately deterministic , and becomes more so the purer the state becomes .
by contrast , with a single measurement and no control ( the parallel measurement case ) , the purification is far from deterministic . in particular
the long tail of low purities means it is impossible to obtain accurate results by replacing @xmath186 by @xmath187 @xcite .
the `` splitting '' of the measurements into `` good '' and `` bad '' is also applicable when the goal of minimizing the mean time @xmath18 is considered . as in the single - detector case @xcite , we take the log - entropy evolution , @xmath189 ; similarly to that case @xmath190 should have more symmetric distribution and therefore its average rate of change would correspond to the mean time of first passage ( mtfp ) . using it equation for three ideal detections ,
the average change of log - entropy in the high - purity limit ( @xmath191 ) reads : @xmath192 i.e. , the three detector measurements now `` splits '' into one `` good '' measurement ( @xmath182 ) , directed along the state , which gives the rate @xmath193 as in eq .
( [ time_pure1 ] ) [ wiseman and ralph protocol ] , and two `` bad '' measurements ( @xmath180 , @xmath181 ) , directed perpendicular to the state , that give one - half of this rate each , see eq .
( [ tau_perp ] ) .
correspondingly , the mean time @xmath194 for the average log - entropy @xmath195 to reach @xmath104 is given by @xmath196 this is two times shorter than the mean time for a single - detector no - control measurement , @xmath197 , eq .
( [ time_pure1 ] ) .
therefore , in comparison to the parallel single measurement case , three complementary measurements give a speed - up of 2 in terms of the mean time to attain a given purity .
we will verify this result by explicitly calculating the mtfp , in the following subsection where we consider non - ideal detectors .
it is now interesting to ask the question : given non - ideal detectors with @xmath198 , what level of purity , @xmath199 can be reached and what time is needed ?
we now consider in detail this question , with emphasis on the high - purity and high - efficiency limit , when @xmath116 and @xmath200 .
the answer to the above question will depend on the goal examined under purification . since the average purity @xmath202 described by the fpe is numerically close to that from the naive ensemble - average evolution eq .
( [ av - pur - rate - naive ] ) , @xmath203 ( see also ref.@xcite ) , we will use the latter in our analysis .
in particular , the stationary value @xmath204 is close to the true stationary value , obtained from the stationary distribution , eq .
( [ fp - pur - stationary ] ) , and reads : @xmath205 in the high - ideality limit it gives @xmath206 .
therefore , @xmath17 can reach a purity level @xmath201 only if @xmath207 the time for @xmath17 to reach @xmath3 can be calculated from eq .
( [ av - pur - rate - naive ] ) , and in the high - purity limit it gives : @xmath208 \label{p - tau - pure - delta } .\ ] ] here we note that the purification rate can be again understood by splitting into `` good '' and `` bad '' measurements .
indeed , the single detector average purification for @xmath78 can be calculated from eq .
( [ bloch-1-det ] ) : @xmath209 and one can observe that @xmath210 is maximized for a measurement in perpendicular direction ( @xmath211 ) for not too small @xmath65 ( e.g. even for a totally mixed state , @xmath156 , this happens for @xmath212 ) .
thus , the first term in eq .
( [ av - pur - rate ] ) comes exactly from two `` good '' measurements ( perpendicular to the state ) , @xmath213 , while the second term comes from the `` bad '' measurement ( parallel to the state ) as in eq .
( [ ito - dp0 ] ) , and is suppressed in the high - purity limit . therefore , the asymptotic result for @xmath214 comes entirely from the two good measurements , as in the ideal case . the time increase in eq .
( [ p - tau - pure - delta ] ) with respect to the ideal measurement , eq .
( [ 3det - tau - pur ] ) is linear in the ratio @xmath15 for @xmath215 and diverges logarithmically to @xmath216 as the impurity level @xmath217 approaches the bound , eq .
( [ p - bound ] ) . to reach impurity level @xmath218
is simply impossible .
we now proceed with the investigation of the mean time @xmath18 for the purity to reach @xmath99 for three complementary non - ideal measurements . following ref .
@xcite we write the it equation for the log - entropy : @xmath221 similar to the ideal case , the average change of log - entropy can be represented as a sum of the measurement along the state and two measurements perpendicular to the state .
indeed , the change @xmath222 for a single non - ideal detector can be written as : @xmath223 we note that for given entropy @xmath224 , the average change @xmath225 is maximized again for @xmath110 and @xmath226 , and it coincides with the ideal case : @xmath227 , see eq .
( [ time_pure1 ] ) . on the other hand
the change due to measurement in the complementary directions ( @xmath228 ) is deterministic and read : @xmath229 $ ] , so that @xmath230 . by considering highly ideal detectors , @xmath231 , so that the last , singular , term in @xmath232 can be neglected
, one can reproduce the mean time @xmath18 by integration of the equation for @xmath233 .
however , in the high - purity limit this is not possible : naive integration of log - entropy , with the singular term , @xmath234 , included , will lead to a wrong result . since in the high - purity limit @xmath200 , @xmath235 , this term is of the order of @xmath236 ,
it is clear that the time of reaching a certain purity level will be essentially affected when @xmath237 . in what follows
, we investigate the exact solution for the mtfp in case of inefficient detectors .
we consider an initially completely mixed state , and denote the mtfp as @xmath238 .
the fpe constructed above , based on the stochastic it eq .
( [ ito - dp-3det ] ) is used , where in the high - purity limit , we approximate the coefficient , eq .
( [ fp - b - coeff ] ) , as : @xmath239 , while keeping @xmath240 unchanged .
the mtfp solution for one reflective boundary ( at @xmath156 ) and one absorptive ( at @xmath241 ) is written as ( see , e.g. ref .
@xcite ) @xmath242 where @xmath243 is the initial purity as before ; in what follows we consider only @xmath244 .
the function that enters the solution is @xmath245 and in our case one obtains @xmath246^{\frac{1}{(1-\delta ) } } } \label{psi - mtfp } , \end{aligned}\ ] ] where we denoted @xmath247 . for ideal measurements , @xmath248
, one can integrate eq .
( [ mtfp ] ) to obtain analytically the result : @xmath249 in agreement @xcite with the straightforward integration of the ensemble - averaged log - entropy equation , [ see eqs.([ito - dlog_s3 ] ) and ( [ mean - time-3det ] ) above ] . for @xmath250 analytical integration is not as obvious : naive expansion of the integrand in series in @xmath95 leads to a sum of contributions with increasing singularities .
we present instead numerical calculations that show the scaling behavior in the high - purity limit .
the main observation is that in the high - purity limit the time of purification can be represented as @xmath251 that is , the time increase @xmath252 depends only on the ratio @xmath253 .
it can be shown that for @xmath254 , @xmath255 , while for @xmath256 , @xmath257}$ ] , with @xmath258 , approaching 0.5 for large @xmath259 . , for small @xmath260 , and @xmath116 ; scaling is established for @xmath261.,width=355 ] \approx c_1 a$ ] , for large @xmath256 , and @xmath116 ; scaling is for @xmath261.,width=374 ] on figs .
[ fig3 ] and [ fig4 ] we present evidence of scaling .
3 shows @xmath262 at fixed @xmath217 and for @xmath260 .
the curve for @xmath263 is close to the curves for @xmath264 that practically coincide on the figure ; the slope at @xmath265 is @xmath266 .
4 shows the scaling for large @xmath259 .
the @xmath267 $ ] , eventually approaches a single curve for @xmath116 .
for @xmath268 it looks like a linear growth , i.e. @xmath267 \approx c_1 a$ ] .
more precisely , by numerical calculation of the derivative @xmath269/da $ ] ( not shown ) one can see that it is not a constant but varies somewhat as discussed above . for small enough @xmath217 ( @xmath270 ) , @xmath271 approaches @xmath272 .
thus , the numerical calculations support eq .
( [ scaling ] ) .
in this paper we investigate in detail a recently proposed quantum state purification protocol based on simultaneous continuous measurement of the three complementary observables for a qubit : @xmath0 , @xmath1 , @xmath2 .
contrary to analogous single - detector purification protocols , that are based on complementarity @xcite , and which require feedback control , here there is no need of introducing quantum feedback .
however , interestingly enough , the purification dynamics of our protocol can be understood via the above mentioned protocols , using the established isotropy of the qubit evolution under three complementary measurements by identical detectors @xcite . for ideal ( i.e. quantum - limited , or efficient ) measurements ,
our main results are the observation of different factors for the speed - up ( relative to a single measurement in the eigenbasis of the qubit state ) , of 4 and 2 , depending on how the purification speed is quantified . in the first case it is in terms of the time when the average system purity , @xmath273 , reaches certain purity level @xmath3
; in the second case it is in terms of the mean time @xmath98 at which the purity @xmath219 first attains the set level of @xmath3 .
both of these speed - ups can be understood via the possibility to `` split '' the three detector measurement at each measurement time step into three equivalent measurements one parallel to the state ( in the bloch sphere sense ) and the remaining two perpendicular to the state which is one of the consequences of isotropy of the qubit evolution in the bloch space .
for the first measure of purification time these correspond to one `` bad '' and two `` good '' measurements ( giving speed - up contributions of @xmath274 and @xmath275 respectively , totalling 4 ) . for the second measure , they correspond to one `` good '' and two bad measurements ( giving speed - up contributions of @xmath276 and @xmath277 respectively , totalling 2 ) . for a measurement with non - ideal detectors ,
the dynamics remains isotropic in the bloch space .
moreover , the classification to `` good '' and `` bad '' measurements remains the same , as long as the detector inefficiency @xmath95 is not greater than @xmath272 .
the inefficiency causes an increase in the purification times , that scales as a function of the ratio of inefficiency over impurity , @xmath15 , in the high - purity and high - efficiency limit , @xmath200 , @xmath116 . here
, the first speed - up quantification , stated in terms of attaining a certain average purity , is simply impossible if @xmath218 , and the time required diverges as @xmath217 approaches this limit from above . on the other hand , the second quantification , stated in terms of the mean time for individual ( stochastically evolving ) systems to attain this purity , always gives an answer .
however we show that the mean time increases exponentially : @xmath278}$ ] for very large @xmath15 ratio .
still , for moderate @xmath279 the mean time is not too much greater than the ideal case estimation .
the authors thank alexander n. korotkov for useful discussions .
rr was supported by lundbeck foundation .
km was supported by the eu integrated project aqute .
hmw and jc were supported by the australian research council centre of excellence ce110001027 .
jc also acknowledges support from national science foundation grant no .
phy-0903953 & phy-1005540 , as well as office of naval research grant no .
n00014 - 11 - 1 - 008 .
r. ruskov and a. n. korotkov , phys .
b * 66 * , 041401(r ) ( 2002 ) ; q. zhang , r. ruskov , and a. n. korotkov , phys .
b * 72 * , 245322 ( 2005 ) ; a. n. korotkov , phys . rev .
b * 71 * , 201305(r ) ( 2005 ) .
r. ruskov and a. n. korotkov , phys .
b * 67 * , 241305(r ) ( 2003 ) ; w. mao , d. v. averin , r. ruskov , and a. n. korotkov , phys . rev . lett . * 93 * , 056803 ( 2004 ) .
c. ahn , a. c. doherty , and a. j. landahl , phys .
a , * 65 * , 042301 ( 2002 ) ; c. ahn , h. m. wiseman , and g. j. milburn , _ ibid . _
* 67 * , 052310 ( 2003 ) ; m. sarovar , c. ahn , k. jacobs , and g. j. milburn , _ ibid . _
* 69 * , 052324 ( 2004 ) .
an important difference is that a classical bayesian inference reveals a pre - existing system state ( e.g. a `` true '' probability distribution ) , while quantum evolution , eqs .
( 1 ) , ( 2 ) , is about the ( final ) quantum state available to an experimentalist . in the rotated frame @xmath280 of fig .
[ fig2 ] , the joint distribution of wiener noises is still _ the same _
@xmath281 gaussian distribution , since the measurements commute for small times . using the actual fpe coefficient @xmath282 , eq .
( [ fp - b - coeff ] ) , numerical integration of eq .
( [ mtfp ] ) gives for the mtfp , @xmath283 , which differs negligibly from eq.([mean - time-3det ] ) for large @xmath284 . | we consider qubit purification under simultaneous continuous measurement of the three non - commuting qubit operators @xmath0 , @xmath1 , @xmath2 .
the purification dynamics is quantified by ( i ) the average purification rate , and ( ii ) the mean time of reaching given level of purity , @xmath3 . under ideal measurements ( detector efficiency @xmath4 ) , we show in the first case an asymptotic mean purification speed - up of @xmath5 as compared to a standard ( classical ) single - detector measurement . however by the second measure
the mean time of first passage @xmath6 of the purity the corresponding speed - up is only @xmath7 .
we explain these speed - ups using the isotropy of the qubit evolution that provides an equivalence between the original measurement directions and three simultaneous measurements , one with an axis aligned along the bloch vector and the other with axes in the two complementary directions . for inefficient detectors ,
@xmath8 the mean time of first passage @xmath9 increases since qubit purification competes with an isotropic qubit dephasing . in the asymptotic high - purity limit ( @xmath10 )
we show that the increase possesses a _ scaling behavior _ : @xmath11 is a function only of the ratio @xmath12 .
the increase @xmath13 is linear for small argument but becomes exponential @xmath14 for @xmath15 large . |
the discovery of superconductivity in sr@xmath0ruo@xmath1 by maeno and coworkers @xcite has attracted widespread attention partially because of the structural similarity with the cuprate - based high - temperature superconductors ( htsc s ) . while the mother compounds of the htsc s are charge transfer insulators , and they usually need to be doped to become metallic and show superconductivity , in sr@xmath0ruo@xmath1 low - temperature superconductivity at @xmath7 = 1.5 k condenses from a metallic state that is a strongly two - dimensional fermi liquid below about 20 k. the fermi surface consists of three weakly corrugated cylindrical sheets @xmath8 being hole - like and @xmath9 and @xmath10 which are electon - like @xcite .
this is in contrast to la@xmath0cuo@xmath1 where only one fermi sheet exists .
the mechanism of conduction in sr@xmath0ruo@xmath1 at higher temperatures is an interesting issue because the @xmath3-axis resistivity shows a broad maximum at around 130 k and for increasing temperatures the resistivity starts to decrease . in @xcite
such a metallic to nonmetallic crossover in the @xmath3-axis resistivity of a highly anisotropic metal like sr@xmath0ruo@xmath1 with no corresponding feature in the @xmath11-plane properties has been related to the strong coupling between the electrons and a bosonic mode propagating and polarized in @xmath3-direction , like the axial oxygen breathing mode @xmath12 , at the @xmath13 point of the brillouin zone ( bz ) to be discussed in .
superconductivity in sr@xmath0ruo@xmath1 is thought to be of unconventional character and thus is intensively discussed in the literature .
the current state of the experiments points to triplet superconductivity in sr@xmath0ruo@xmath1 in contrast to the @xmath14-wave singlet pairing in the htsc s .
most discussion of the superconducting mechanism for sr@xmath0ruo@xmath1 is focussed on magnetic fluctuations and coulomb repulsion , for a review see @xcite .
however , a contribution of electron - phonon interaction ( epi ) to pairing may not be ruled out because the superconducting transition exhibits a clear isotope effect @xcite . in case of the htsc s
there is increasing evidence that epi is strong and phonons might play an important role for the electron dynamics , see @xcite , @xcite and references therein .
thus , a detailed theoretical study of phonon dynamics in sr@xmath0ruo@xmath1 and a comparison with the situation in the htsc s is desirable .
the low - temperature superconductor sr@xmath0ruo@xmath1 crystallizes in the k@xmath0nif@xmath1 structure ( i4/mmm ) isostructural to the htsc la@xmath0cuo@xmath1 .
an analysis and comparison of the lattice dynamics in both materials is the main intention of the present work . for this purpose
we compute the phonon dynamics for sr@xmath0ruo@xmath1 and oppose it with our earlier calculations for la@xmath0cuo@xmath1 .
the most characteristic feature of phonon dynamics in the htsc s is the strong softening upon doping of the high - frequency oxygen bond - stretching modes ( obsm ) which seems to be generic for @xmath4- as well as @xmath5-doped cuprates , see @xcite .
the strong frequency renormalization and a related increase of the linewidths observed for these anomalous modes point to a strong coupling of the phonons to the charge dynamics . a discussion and theoretical investigations of these phonon anomalies
are presented e.g. in @xcite . in agreement with inelastic neutron scattering @xcite our calculations demonstrate that the typical obsm phonon anomalies of the cuprates are not present in sr@xmath0ruo@xmath1 .
moreover the characteristic @xmath6 branch related to the strongly coupling @xmath12 mode in la@xmath0cuo@xmath1 is missing . in order to achieve reliable results for the @xmath6 phonon modes propagating and polarized along the @xmath3-axis a careful calculation of
the proper polarization part of the density response function is crucial .
this has already been shown for la@xmath0cuo@xmath1 in @xcite where we have proven that the @xmath6 modes are highly sensitive with respect to the charge response perpendicular to the cuo plane .
the large anisotropy of the electronic structure of the cuprates is as a rule considerably underestimated in dft - lda calculations and as a consequence the @xmath6 modes are not well described .
so we have modified a lda - based calculation for la@xmath0cuo@xmath1 to account for the much weaker @xmath2-dispersion of the electronic bandstructure in the real material .
we have optimized the interlayer coupling in such a way that the significant features of the sensitive @xmath6 phonons are well described .
the same route is followed in the present work for sr@xmath0ruo@xmath1 to get a realistic representation of the @xmath6 modes and in particular of @xmath12 which can not be described in the ionic shell - model taking additionally homogeneous electron gas screening into account @xcite .
finally , we investigate the question if there is room for a phonon - plasmon scenario around the @xmath3-axis for sr@xmath0ruo@xmath1 which has been shown to be a realistic option in la@xmath0cuo@xmath1 @xcite . due to
the much weaker electronic @xmath2-dispersion obtained in our computations as compared to la@xmath0cuo@xmath1 , i.e. a much stronger anisotropy of sr@xmath0ruo@xmath1 , the calculated free - plasmon frequencies along the @xmath3-axis are about a factor of eight smaller in the collisionless regime .
damping , generated by interactions between the electrons as well as interband transitions which are on a much lower energy scale as for la@xmath0cuo@xmath1 very likely leads to overdamping of the plasmons propagating strictly along the @xmath3-axis in sr@xmath0ruo@xmath1 .
thus , different from la@xmath0cuo@xmath1 coupled @xmath3-axis phonon - plasmon modes should not be well defined collective excitations at least strictly along the @xmath15 direction and accordingly do not enter the list of possible players for pairing in contrast to la@xmath0cuo@xmath1 .
the article is organized as follows . in the theory necessary to understand the calculated results
is shortly reviewed .
presents the calculations .
the modification of a lda - based electronic bandstructure providing the basis of the single particle content of the irreducible polarization part of the density response function is developed .
furthermore , the phonon dispersion is calculated in adiabatic approximation for sr@xmath0ruo@xmath1 and a comparative discussion with the situation in la@xmath0cuo@xmath1 is given . finally , in a nonadiabatic calculation the possibility of a phonon - plasmon scenario is examined .
a summary of the paper is given in and the conclusions are drawn .
in the following a brief survey of the theory and modeling is presented .
a detailed description can be found in @xcite and in particular in @xcite where the calculation of the coupling parameters of the theory is presented .
the local part of the electronic charge response and the epi is approximated in the spirit of the quasi - ion approach @xcite by an ab initio rigid ion model ( rim ) taking into account covalent ion softening in terms of ( static ) effective ionic charges calculated from a tight - binding analysis .
the tight - binding analysis supplies these charges as extracted from the orbital occupation numbers @xmath16 of the @xmath17 ( tight - binding ) orbital in question : @xmath18 @xmath19 stands for the @xmath17-component of the eigenvector of band @xmath5 at the wavevector @xmath20 in the first bz ; the summation in runs over all occupied states and @xmath21 gives the number of the elementary cells in the ( periodic ) crystal .
in addition , scaling of the short - ranged part of certain pair potentials between the ions is performed to simulate further covalence effects in the calculation in such a way that the energy - minimized structure is as close as possible to the experimental one @xcite .
structure optimization and energy minimization is very important for a reliable calculation of the phonon dynamics through the dynamical matrix .
taking just the experimental structure data as is done in many cases in the literature may lead to uncontrolled errors in the phonon calculations .
the rim with the corrections just mentioned then serves as an unbiased reference system for the description of the htsc s and can be considered as a first approximation for the insulating state of these compounds .
starting with such an unprejudiced rigid reference system non - rigid electronic polarization processes are introduced in form of more or less localized electronic charge - fluctuations ( cf s ) at the outer shells of the ions . especially in the metallic state of the htsc
s the latter dominate the _
nonlocal _ contribution of the electronic density response and the epi and are particularly important in the cuo planes and the ruo plane . in addition , _ anisotropic _ dipole - fluctuations ( df s ) are admitted in our approach @xcite , which prove to be specifically of interest for the ions in the ionic layers mediating the dielectric coupling and for the polar modes .
thus , the basic variable of our model is the ionic density which is given in the perturbed state by @xmath22 @xmath23 is the density of the unperturbed ion , as used in the rim , localized at the sublattice @xmath8 of the crystal and moving rigidly with the latter under displacement .
the @xmath24 and @xmath25 describe the amplitudes and the form - factors of the cf s and the last term in equation represents the dipolar deformation of an ion @xmath8 with amplitude ( dipole moment ) @xmath26 and a radial density distribution @xmath27 .
@xmath28 denotes the unit vector in the direction of @xmath29 .
the @xmath25 are approximated by a spherical average of the orbital densities of the ionic shells calculated in lda taking self - interaction effects ( sic ) into account . the dipole density @xmath30 is obtained from a modified sternheimer method in the framework of lda - sic @xcite .
all sic - calculations are performed for the average spherical shell in the orbital - averaged form according to perdew and zunger @xcite . for the correlation part of the energy per electron
the parametrization given in @xcite has been used .
the total energy of the crystal is obtained by assuming that the density can be approximated by a superposition of overlapping densities @xmath31 .
the @xmath23 in equation are also calculated within lda - sic taking environment effects , via a watson sphere potential and the calculated static effective charges of the ions into account .
the watson sphere method is only used for the oxygen ions and the depth of the watson sphere potential is set as the madelung potential at the corresponding site .
such an approximation holds well in the htsc s @xcite . finally , applying the pair - potential approximation we get for the total energy : @xmath32 the energy @xmath33 depends on both the configuration of the ions @xmath34 and the electronic ( charge ) degrees of freedom ( edf ) @xmath35 of the charge density , i.e. @xmath36 and @xmath37 in equation .
@xmath38 are the energies of the single ions .
@xmath39 , @xmath40 denote the elementary cells and @xmath8 , @xmath9 the corresponding sublattices .
the second term in equation is the interaction energy of the system , expressed in terms of _ anisotropic _ pair - interactions @xmath41 .
both @xmath38 and @xmath41 in general depend upon @xmath42 via @xmath31 in equation .
the pair potentials in equation can be seperated into long - ranged coulomb contributions and short - ranged terms , for details see e.g. ref @xcite .
the dynamical matrix in harmonic approximation can be derived as @xmath43_{\rm rim}\\ & - \frac{1}{\sqrt{m_\alpha m_\beta } } \sum_{\kappa,\kappa ' } \left[b^{\kappa\alpha}_i({\mathbf{q } } ) \right]^ { * } \left[c^{-1}({\mathbf{q } } ) \right]_{\kappa\kappa ' } b^{\kappa'\beta}_j({\mathbf{q}}).\end{aligned}\ ] ] the first term on the right hand side denotes the contribution from the rim .
@xmath44 , @xmath45 are the masses of the ions and @xmath46 is a wave vector from the first bz .
the quantities @xmath47 and @xmath48 in equation represent the fourier transforms of the electronic coupling coefficients and are calculated from the energy in equation , i.e. @xmath49 @xmath50 denotes the edf ( cf and df in the present model , see equation ) in an elementary cell .
the @xmath51 coefficients describe the coupling between the edf and the displaced ions ( bare electron - phonon coupling ) , and the coefficients @xmath52 determine the interaction between the edf . the phonon frequencies @xmath53 and the corresponding eigenvectors @xmath54 of the modes @xmath55 are obtained from the secular equation for the dynamical matrix in equation , i.e. @xmath56 the eqs
. - are generally valid and , in particular , are independent of the specific model for the decomposition of the perturbed density in equation and the pair approximation in equation for the energy .
the lenghty details of the calculation of the coupling coefficients @xmath51 and @xmath52 can not be reviewed in this paper .
they are presented in @xcite . in this context
we remark that the coupling matrix @xmath57 of the edf - edf interaction , whose inverse appears in equation for the dynamical matrix , can be written in matrix notation as @xmath58 @xmath59 is the inverse of the _ irreducible ( proper ) polarization part _ of the density response function ( matrix ) and contains the kinetic part to the interaction @xmath52 while @xmath60 embodies the hartree and exchange - correlation contribution .
@xmath61 needed for the dynamical matrix and the epi
is closely related to the ( linear ) density response function ( matrix ) and to the inverse dielectric function ( matrix ) @xmath62 , respectively .
only very few attempts have been made to calculate the phonon dispersion and the epi of the htsc s using the linear response method in form of density functional perturbation theory ( dfpt ) within lda @xcite .
these calculations correspond to calculating @xmath63 and @xmath60 in dft - lda and for the _ metallic _ state only . on the other hand , in our microscopic modeling dft - lda - sic calculations
are performed for the various densities in equation in order to obtain the coupling coefficients @xmath51 and @xmath60 .
including sic is particularly important for localized orbitals like cu3d in the htsc s .
our theoretical results for the phonon dispersion @xcite , which compare well with the experiments , demonstrate that the approximative calculation of the coupling coefficients in our approach is sufficient , even for the localized cu3d states .
written in matrix notation we get for the density response matrix the relation @xmath64 the cf - cf submatrix of the matrix @xmath63 can approximatively be calculated for the metallic ( but not for the undoped and underdoped ) state of the htsc s from a tba of a single particle electronic bandstructure . in this case
the electronic polarizability @xmath63 in tight - binding representation reads : @xmath65 \left[c_{\kappa ' n}^{*}({\mathbf{k}})c_{\kappa ' n'}({\mathbf{k}}+{\mathbf{q } } ) \right]^{*}.\end{aligned}\ ] ] @xmath66 , @xmath33 and @xmath52 in equation are the occupation numbers , the single - particle energies and the expansion coefficientes of the bloch - functions in terms of tight - binding functions .
the self - consistent change of an edf at an ion induced by a phonon mode @xmath67 with frequency @xmath53 and eigenvector @xmath54 can be derived in the form @xmath68 e^{i{\mathbf{q}}{\mathbf{r}}_\kappa^{\mathbf{a}}}\nonumber\\ & \equiv & \delta\zeta_\kappa({\mathbf{q}}\sigma)e^{i{\mathbf{q}}{\mathbf{r}}^{\mathbf{a}}},\end{aligned}\ ] ] with the displacement of the ions @xmath69 the self - consistent response per unit displacement of the edf in equation is calculated in linear response theory as : @xmath70 the generalization for the quantity @xmath63 in eqs . and needed for the kinetic part of the charge response in the nonadiabatic regime , where dynamical screening effects must be considered , can be achieved by adding @xmath71 to the differences of the single - particle energies in the denominator of the expression for @xmath63 in equation .
other possible nonadiabatic contributions to @xmath52 related to dynamical exchange - correlation effects and the phonons themselves are beyond the scope of the present approach . using equation for the dielectric matrix , @xmath72 , and the frequency - dependent version of the irreducible polarization part , @xmath63 , according to equation , the free - plasmon dispersion is obtained from the condition , @xmath73 = 0 .\ ] ] the coupled - mode frequencies of the phonons and the plasmons must be determined self - consistently from the secular equation for the dynamical matrix which now contains the frequency @xmath74 implicitly via @xmath63 in the response function @xmath61 .
such a nonadiabatic approach is necessary for a description of the interlayer phonons and the charge - response within a small region around the @xmath3-axis as performed in @xcite and in the present paper for sr@xmath75ruo@xmath76 .
cuo@xmath1 of the @xmath3-axis polarized @xmath77 modes @xcite based on the 31bm for the proper polarization part @xmath78 ( a ) and the m31bm ( b ) , respectively .
the experimental data are represented as open squares ( @xmath79 ) .
the full dot @xmath80 denotes the @xmath81 mode and the open circle ( @xmath82 ) the @xmath83(ferro ) mode . ]
we shortly recall the construction of a modified lda - based model for la@xmath0cuo@xmath1 which better describes the real anisotropy of the htsc s that as a rule is not correctly represented in lda being generally too isotropic . in our effort to obtain a reliable description of the @xmath6 phonons polarized and propagating along the @xmath3-axis of the cuprates
@xcite we have shown that these modes are highly sensitive with respect to details of the @xmath3-axis coupling and thus a very accurate representation of the electronic @xmath2-dispersion is needed .
typical lda based computations lead to an overestimation of the @xmath2-dispersion and give imprecise results for the @xmath6 modes .
in we illustrate the degree of inaccuracy for the @xmath6 phonons , in la@xmath0cuo@xmath1 as obtained in @xcite .
( a ) shows the result of an lda - based tight - binding representation of the electronic bandstructure ( 31 band model , 31bm ) and ( b ) gives the outcome of a modified bandstructure ( m31bm ) with reduced first neighbour @xmath84-la parameters of the 31bm by 1/6 and of the first neighbour la - la parameters by 1/3 . as seen from this
leads to a much better result for the phonon modes .
the characteristic experimental features of the dispersion i.e. the step - like structure of the second highest branch and most significant the third highest branch with the steep dispersion towards the @xmath13 point are not well reflected by the calculation with the 31bm as input for the proper polarization part in equation .
using , however , the m31bm the calculated phonon dispersion is in good agreement with the experiment as can be seen in ( b ) .
the characteristic features are now well described .
we also find as a consequence of the increased anisotropy in the m31bm a rearrangement of the three @xmath13 point modes with the highest energy . while in the 31bm
the strongly coupling @xmath12 mode is the second highest mode and the @xmath85(ferro ) mode the lowest one in the m31bm @xmath12 is the lowest and @xmath86(ferro ) the second highest of the three modes .
thus @xmath12 is the end point of the steep branch . during this mode rearrangement
the frequency of @xmath12 stays nearly constant while that of @xmath87 ( ferro ) is strongly increased .
we have recalled these facts for la@xmath0cuo@xmath1 because for sr@xmath0ruo@xmath1 a comparable high sensibility of the @xmath6 modes in dependence of the electronic @xmath2-dispersion occurs and an analogous mode behaviour for @xmath12 and @xmath85(ferro ) results in response to a weaker @xmath2-dispersion .
likewise as in la@xmath0cuo@xmath1 this sensitivity of certain @xmath3-axis phonons is used to construct an accurate tight - binding description of the electronic bandstructure ( bs ) in sr@xmath0ruo@xmath1 . as a first approximation for the bs we employ a tight - binding representation of the first principles linearized - augmented - plane - wave ( lapw ) bs as obtained within the framework of dft - lda @xcite .
this analysis leads to a 27 band model ( 27bm ) including sr4d , ru4d and o2p states .
ruo@xmath1 of the @xmath6 modes based on different models for the electronic bandstructure as explained in the text .
27bm ( a ) , m27bm1 ( b ) and m27bm2 ( c ) .
the experimental values from @xcite are represented as open squares ( @xmath79 ) .
the full dot @xmath80 denotes the @xmath81 mode and the open circle @xmath88 the @xmath83(ferro ) mode . ] ruo@xmath1 of a @xmath89 mode ( @xmath90 ) and a @xmath91 mode ( @xmath92 ) based on the 27bm ( a ) , the m27bm1 ( b ) and the m27bm2 ( c ) .
the experimental data points @xcite are characterized by open squares ( @xmath79 ) for @xmath89 and by full squares @xmath93 for @xmath91 .
the full dot @xmath80 and the open circle ( @xmath82 ) at the @xmath13 point represent @xmath12 and @xmath87 , respectively . ]
the associated calculated phonon dispersion for the @xmath6 modes of sr@xmath0ruo@xmath1 along the @xmath15 direction is depicted in ( a ) and for the relevant @xmath89 and @xmath91 mode along the @xmath94 direction in ( a ) together with the experimental data points @xcite .
figures [ fig_sro_lam](b ) , ( c ) and figures [ fig_sro_del1_4](b ) , ( c ) , respectively , display our results for two modified 27 band models ( m27bm1 and m27bm2 ) with the tight - binding parameters important for @xmath3-axis is coupling ( o@xmath95-o@xmath96 , ru - sr , o@xmath95-sr , ru - o@xmath96 ) reduced by 1/2 and 1/5 with respect to the lda - based 27bm .
this leads as shown in figures [ fig_efsurf_kz](a)-(c ) to a much stronger anisotropy as in the 27bm , i.e. to a strongly reduced @xmath2-dispersion which is also much weaker than in la@xmath0cuo@xmath1 @xcite .
-dispersion for sr@xmath0ruo@xmath1 along the cut of the fermi surface highlighted in ( a ) by the black bar .
27bm ( a ) , m27bm1 ( b ) and m27bm2 ( c ) .
the corners of the surfaces are @xmath97 , @xmath98 , @xmath99 , @xmath100 . ]
ruo@xmath1 in the @xmath101 plane in model m27bm2 ( a ) and the 27bm ( b ) with the measured fermi surface from @xcite . the open dots ( @xmath82 )
give the experimental results .
for the explanation of the small black bar in ( a ) see . ] of sr@xmath0ruo@xmath1 in the m27bm2 taking into account the strongly enhanced anisotropy of the real material as compared with the 27bm as a typical dft - lda based model . ] in the reduction of the electronic @xmath2-dispersion is shown along the cut of the fermi surface ( fs ) highlighted in figure [ fig_efsurf_xy](a ) when going from the 27bm to the most anisotropic m27bm2 via an intermediate @xmath3-axis coupling in m27bm1 .
the best result for the phonon modes in figures [ fig_sro_lam ] and [ fig_sro_del1_4 ] , which are less well described in the 27bm , are obtained for the highly anisotropic m27bm2 , i.e. a nearly two - dimensional bs reflected by an even considerably weaker @xmath2-dispersion as in case of la@xmath0cuo@xmath1 , see and @xcite .
it is enlightening to point out that in context with the @xmath89 and @xmath91 branches in an investigation of @xcite exhibits that the dispersion can not be described within a normal ionic shell model extended by homogeneous electron gas screening to simulate the metallic character of sr@xmath0ruo@xmath1 .
two specially adapted force constants between ru - o@xmath96 and ru - o@xmath95 had to be introduced to mimic the observed dispersion .
of course these force constants do not have an intrinsic physical meaning . on the other hand , from our calculation
we can conclude that on a microscopic level an accurate electronic dispersion along the @xmath3-axis is essential to understand the observed mode behaviour .
ruo@xmath1 . from left
to right we display in the first row @xmath102(ferro ) , @xmath103 , @xmath104 ; in the second row @xmath105 , @xmath12 , @xmath85(ferro ) and in the third row @xmath106 and @xmath107 . ] for a more detailed discussion of the rearrangement of the phonon modes in figures [ fig_sro_lam ] and [ fig_sro_del1_4 ] as a result of a modification of the @xmath2-dispersion of the bs it is useful to display in the displacement patterns of some phonon modes relevant for our studies . in the 27bm
the frequency of @xmath12 is nearly degenerate with @xmath85(ferro ) .
we denote the latter mode as `` ferroelectric '' because its displacement pattern looks similar to that of the ferro - like @xmath102(ferro ) mode at the @xmath108-point , see . here
the oxygen anions vibrate coherently against the cations in the lattice and as a consequence the electric dipole moments generated by the motion add constructively . with decreasing @xmath2-dispersion of the bs in m27bm1 and m27bm2 achieved by a reduction of the relevant tight - binding parameters mentioned above by 1/2 and 1/5 , respectively ,
the frequency of @xmath85(ferro ) increases by about 2 thz while @xmath12 remains virtually unchanged .
this highly sensible behaviour with regard to @xmath3-axis coupling in sr@xmath0ruo@xmath1 has also been demonstrated for la@xmath0cuo@xmath1 @xcite where the frequency of @xmath85 ( ferro ) increases by more than 3 thz as can bee read off from .
@*9l & & + & ru4d & o@xmath952p & o@xmath962p & sr4d & ru4d & o@xmath952p & o@xmath962p & sr4d m27bm & 26.98 & 7.62 & 0.02 & 0.49 & 0 & 0 & 3.22 & 0.34 + 27bm & 27.34 & 6.43 & 0.63 & 3.57 & 0 & 0 & 9.19 & 4.69 the physical origin for the sensitivity derives from the fact that for symmetry reasons in case of @xmath85(ferro ) cf s can be excited in the screening process only at the ions in the ionic layers , i.e. o@xmath96 , la in la@xmath0cuo@xmath1 and o@xmath96 , sr in sr@xmath0ruo@xmath1 . the strength of these cf s on the other hand is governed by the matrix elements of the proper polarization part @xmath78 in equation of the corresponding out - of - plane ions and , thus , depends critically on the coupling along the @xmath3-axis .
contrarily , the renormalization of @xmath12 is nearly exclusively given by the cf s on the ru4d and o@xmath952p orbitals in the plane .
for quantitative results compare with the cf s listed in .
this explains the inertness of @xmath12 and the sensibility of @xmath85(ferro ) with respect to the change of the electronic @xmath2-dispersion . as a further example for the critical interrelation of @xmath3-axis phonon dynamics and electronic @xmath3-axis coupling we note that the steep branch connecting @xmath102(ferro ) with @xmath12 in la@xmath0cuo@xmath1 is missing in sr@xmath0ruo@xmath1 . in the latter case due to mode rearrangement @xmath102(ferro ) as the lowest of the three highest modes at @xmath108 is connected to @xmath109 that is the lowest of the three highest modes at @xmath13 and not to @xmath12 being the second highest mode .
the bs underlying the m27bm2 is illustrated in . as already mentioned the dispersion of the bands along the @xmath110
( @xmath2 ) direction is extremely small , considerably smaller than for the 27bm and for the case of la@xmath0cuo@xmath1 .
thus sr@xmath0ruo@xmath1 is a nearly two - dimensional fermi liquid .
however , as our calculation in figures [ fig_sro_lam ] and [ fig_sro_del1_4 ] have shown the remaining weak three - dimensionality is crucial to obtain a reliable description of certain @xmath3-axis phonons and the @xmath3-axis charge response in general .
strong hybridization between ru4d and o@xmath952p states is evident from the bs in because of the striking in - plane dispersion of the bands .
moreover , we find a similar shape of the partial density of states ( pdos ) of the ru4d and the o@xmath952p orbitals around the fermi energy @xmath111 .
the states at @xmath111 are more strongly of ru4d type with some admixture of o@xmath952p .
the contribution of o@xmath962p and sr4d around @xmath111 is very small . in the m27bm2
we obtain for the pdos at @xmath111 @xmath112 ev@xmath113 , @xmath114 ev@xmath113 , @xmath115 ev@xmath113 and @xmath116 ev@xmath113 .
there are qualitative differences concerning the origin of hybridization comparing sr@xmath0ruo@xmath1 with la@xmath0cuo@xmath1 .
hybridization in sr@xmath0ruo@xmath1 is strongly favoured because the radial extent of the 4d wave functions in ru is much larger than for the 3d wave functions in cu .
concurrently this leads to a weaker influence of electron correlation effects , e.g. a smaller on - site coulomb repulsion @xmath117 for ru4d compared to cu3d . in our calculations we find for @xmath117(cu3d ) = 1.005 dry and for @xmath117(ru4d ) = 0.619 dry . on the other hand ,
hybridization is promoted in la@xmath0cuo@xmath1 as compared to sr@xmath0ruo@xmath1 because the energy levels of the copper and oxygen ions are close by in la@xmath0cuo@xmath1 but far off between ruthenium and oxygen in sr@xmath0ruo@xmath1 .
a further remarkable difference of the bs in sr@xmath0ruo@xmath1 and la@xmath0cuo@xmath1 can be read off from and the bs for la@xmath0cuo@xmath1 in @xcite . in sr@xmath0ruo@xmath1 up to three bands close to each other are crossing the fermi level giving rise to a three - sheet fs while in la@xmath0cuo@xmath1 only one band is crossing and other bands are more far away .
thus we have a completely different situation as _ interband transitions _ are concerned which are on a much lower energy scale in sr@xmath0ruo@xmath1 as in la@xmath0cuo@xmath1 .
so we find for @xmath118 with @xmath119 for @xmath46 at the @xmath13 point in case of sr@xmath0ruo@xmath1 a minimum of 20.67 mev in the m27bm2 around @xmath120 , see also , while in la@xmath0cuo@xmath1 we obtain for the @xmath13 point at @xmath121 a minimum of 452.71 mev .
this , of course , has important consequences for the contribution of the interband transitions to @xmath78 in equation concerning magnitude as well as energy scale . for example
the low energy scale of the interband transitions in sr@xmath0ruo@xmath1 can generate damping of possible plasmons already at very low energy .
such a damping already present in the collisionless regime together with additional damping due to interactions between the electrons very likely leads to overdamping of the very low lying plasmons in sr@xmath0ruo@xmath1 along the @xmath3-axis .
the energy of the latter in the collisionless regime is only 5 - 6 mev along the @xmath110 direction as will be shown in .
thus , the degree of the remaining weak three - dimensionality quantified by our computations is also very important for a possible existence of plasmons and coupled phonon - plasmon modes around the @xmath3-axis .
such modes have been predicted for la@xmath0cuo@xmath1 in our recent calculations @xcite which , however , is not so anisotropic as sr@xmath0ruo@xmath1 and the plasmons along the @xmath3-axis are at higher frequencies that overdamping can be avoided . in figure [ fig_efsurf_xy](a ) and
( b ) , respectively , the calculated fs in the @xmath101 plane is shown for the m27bm2 and the 27bm , together with the experimental data points @xcite .
the fs consists of the sheets @xmath8 , @xmath9 , @xmath10 .
the @xmath8 sheet is holelike and @xmath9 and @xmath10 are electronlike .
the @xmath10 sheet is dominantly derived from the ru4d@xmath95 orbitals , the @xmath8 and @xmath9 sheets are primarily related to the 4d@xmath122 and 4d@xmath123 orbitals of ru which exhibit anticrossing behaviour along the zone diagonal .
the large electronlike @xmath10 sheet passes right through the hybridization gap between the @xmath8 and @xmath9 sheets .
the experimental results are not well described by the 27bm in ( b ) .
in particular the crossing between the @xmath9 and @xmath10 sheet in the 27bm which is typical forlda calculations is experimentally not reproduced .
this is really a crossing and not an anticrossing as for @xmath101 the bands related to d@xmath95 and d@xmath122/d@xmath123 oribitals have different symmetry and no mixing is allowed .
on the other hand , our modified bs model m27bm2 in ( a ) is in good agreement with the experimental fs . in order to obtain also a more global impression of the magnitude of the enhanced anisotropy in the m27bm2
as compared to the 27bm we compare some fs parameters being important for transport like the drude plasma energy tensor and the fermi velocity tensor .
the drude tensor ist defined as @xmath124 and the fermi velocity is given by @xmath125 with @xmath126 @*7l & @xmath127 & @xmath128 & @xmath129 & @xmath130 & @xmath131 & @xmath132 27bm & 1045.94 & 65.50 & 6.63 & 0.88 & 15.97 & 7.53 + m27bm2 & 1017.40 & 3.57 & 6.03 & 0.08 & 284.99 & 75.38 + m31bm & 648.60 & 25.25 & 2.99 & 0.11 & 25.69 & 27.18 the output of our calculation for the 27bm and the m27bm2 is given in .
as can be seen we obtain a large enhancement of the anisotropy ratio for the plasma frequencies @xmath133 and of the fermi velocities @xmath132 by about a factor 18 and 10 , respectively , in the m27bm2 as compared to the lda - based 27bm .
additionally , we have included in the corresponding calculated data for la@xmath0cuo@xmath1 @xcite .
again we recognize the by far larger anisotropy of sr@xmath0ruo@xmath1 also in the transport properties .
the values for @xmath134 and @xmath135 are significantly increased in la@xmath0cuo@xmath1 .
in this section we investigate the phonon dynamics of sr@xmath0ruo@xmath1 in the main symmetry directions @xmath94 , @xmath136 and @xmath15 and continue our discussion from of a comparison of important characteristica as found in our earlier calculations for la@xmath0cuo@xmath1 . for a definite investigation of the phonon renormalization induced by the _ nonlocal _ epi effects of df and cf type mediated by the second term in equation a quantitative reference model for the calculation of the phonon dispersion not including the nonlocal screening effects but representing the important ionic component of binding in the material
is needed . such a model sketched in including approximately the _ local _ epi effects
is provided by the ab initio rim extended via covalent ion softening and scaling of certain short - ranged pair potentials .
ruo@xmath1 for the rim in the main symmetry directions @xmath94 , @xmath136 and @xmath15 . the various symbols representing the experimental results are from @xcite and indicate different irreducible representations .
the arrangement of the panels from left to right according to the different irreducible representations is as follows : @xmath137 , @xmath138 , @xmath139 , @xmath140 . ]
the result of the phonon dispersion along the main symmetry directions is shown for the rim in and compared with the experimental data from inelastic neutron scattering ( ins ) @xcite .
the static effective charges found for the model are ru@xmath141 , o@xmath142 , o@xmath143 and sr@xmath144 .
in particular the charges of ru and o@xmath95 differ considerably from their nominal values ru@xmath145 and o@xmath146 .
this is due to the strong hybridization of the ru4d and o@xmath952p states that reduces the amplitude of the static effective charges in a mixed ionic - covalent compound like sr@xmath0ruo@xmath1 because of the charge transfer ( ct ) from the cations to the anions is not complete as in the entirely ionic case .
consistent with the result of the effective ionic charges is an enhanced covalent character of the ru - o@xmath95 plane with a dominant covalent scaling of the ru - o@xmath95 potential and a more ionic character of the sr - o@xmath96 layer with an ionic scaling of the sr - o@xmath96 potential as found in the calculations .
altogether the rim with these modifications yields good structural data for the energy - minimized configuration . in detail
we get for the planar lattice constant @xmath147 , for the lattice constant along the @xmath3-axis @xmath148 and for the internal position of the o@xmath96 and the sr ion @xmath149 and @xmath150 .
the experimental values at 15 k are @xmath151 , @xmath152 , @xmath153 , @xmath154 @xcite . from these data
we extract that the tetragonal distortion of the ruo@xmath155 octahedra is significantly smaller than for the cuo@xmath155 octahedra in la@xmath0cuo@xmath1 .
the ratio of the in - plane to out - of - plane distance of the oxygen is 0.9196 in sr@xmath0ruo@xmath1 but 0.7706 in la@xmath0cuo@xmath1 .
the phonon dispersion of the rim is in reasonable agreement with the experiment as far as the branches with lower frequencies are concerned .
this points to the importance of the ionic component of binding in this material .
large deviations are observed for the modes with higher frequencies .
in particular the high - frequency obsm @xmath106 , o@xmath156 and @xmath12 , see figure [ fig_evs ] , are not well described .
the corresponding frequencies are overestimated as compared with the full calculation including df s and cf s by about 5.8 thz , 7.4 thz and 6.7 thz , respectively .
this can be attributed according to our approach to the missing screening of the coulomb interaction by df s and most importantly by cf s .
noticeable is the soft @xmath157 mode at the @xmath158 point and the very low frequency of the branch in the measurements ( the notation for the irreducible representations characterizing the symmetry of the modes @xmath157 and @xmath159 is interchanged with the notation of @xcite ) .
this mode is associated with the rotation of ruo@xmath155 octahedra around the @xmath3-axis . as indicated by our calculations
the softness of the rotational mode can be considered as a precursor of a structural phase transition which in the meanwhile has been observed in the ca@xmath160sr@xmath161ruo@xmath1 series @xcite . from our calculation in the rim
this transition can be expected to be driven essentially by the strong component of the ionic forces in sr@xmath0ruo@xmath1 .
contrarily , sr@xmath0ruo@xmath1 does not experience the tilt instability related to the @xmath159 branch at the @xmath158 point found in our calculations for la@xmath0cuo@xmath1 .
this instability correctly indicates the experimentally observed structural phase transition from the high - temperature tetragonal ( htt ) to the low - temperature orthorhombic ( lto ) structure .
also this transition is mainly brought about by the long - ranged ionic forces .
substituting the smaller isovalent ca ions for the sr ions induces a misfit between the ca and ru ions and generates a close connection between the rotation of the ruo@xmath155 octahedra and the electronic and magnetic properties in the ca@xmath160sr@xmath161 ruo@xmath1 series @xcite . besides the softening of the rotational mode in the range @xmath162 around @xmath163 a similar structural transition occurs due to the softening of the ruo@xmath155 tilt mode @xcite . in this context
it is interesting that at @xmath164 in the paramagnetic phase of ca@xmath165sr@xmath166ruo@xmath1 the @xmath10 fermi sheet gains a remarkable @xmath2-dispersion and thus a more three - dimensional character than sr@xmath0ruo@xmath1 @xcite .
this issue also emphasizes an interrelation between the electronic @xmath2-dispersion and the ruo@xmath155 distortions .
ruo@xmath1 . ] in we display the calculated results of the phonon dispersion admitting additionally to the rim anisotropic df s .
this leads to a better agreement of the dispersion curves with the inelastic neutron results . checking these calculations against the rim the width of the spectrum
is reduced towards the experiment .
the frequencies of the transverse and longitudinal optical @xmath167 modes polarized in the plane are decreased by the df s and thereby the largest lo - to splitting is reduced at the @xmath108-point from 8.8 thz in the rim to 6.3 thz when dipolar polarization processes are included . at the same time the frequencies of the @xmath86 modes at @xmath108 and @xmath13 are lowered considerably by the df s in @xmath168-direction and the large splitting of the @xmath102(ferro ) mode of 10.13 thz in the rim is reduced to 8.85 thz .
the high - frequency obsm @xmath106 , @xmath169 and @xmath81 are also decreased due to the screening by the df s but are still overestimated by about 3.9 thz , 5.1 thz and 4 thz as compared with the full calculation displayed in including additionally cf s .
thus , the softening of these modes compared with the rim is of the order 2 - 3 thz and the remaining effect of the renormalization can be assigned to the cf s on the ru4d and o@xmath952p orbitals .
@*6l & ru4d & o@xmath1702p & o@xmath1712p & o@xmath1722p & sr4d @xmath173 & 38.252 & 0 & 4.472 & -0.002 & 0 + o@xmath174 & 50.875 & 0 & 0 & 0.119 & 0.081 + o@xmath175 & 26.984 & 7.621 & 7.621 & 0.022 & 0.486 + o@xmath176 & 12.521 & -3.714 & -3.714 & -2.224 & -0.319 the quantitative results for the cf s according to equation are listed for the obsm in .
the cf s for @xmath104 are also given in the table which prove to be very different from those of @xmath12 despite the same local displacement pattern .
as can be extracted from for @xmath177 and @xmath12 the apex oxygens @xmath178 move locally in phase against or away from the ruo layers .
because of the weak screening along the @xmath3-axis and the corresponding large strength of the nonlocal epi we can expect these vibrations to induce cf s in the ruo layers , see for the calculated results .
however , the cf s differ qualitatively between @xmath12 and @xmath104 , respectively , because for the @xmath108-point vibrations the following sum rule can be derived for the cf s @xmath179 @xcite , @xmath180 mode calculated according to equation for the m27bm2 .
the units are in @xmath181 .
the phase of @xmath12 is as in figure [ fig_evs ] .
full lines ( @xmath90 ) indicate accumulation of electrons in the corresponding region of space and broken lines ( @xmath182 ) indicate regions of repulsion for the electrons . ]
the sum over @xmath50 in equation runs over the cf s in an elementary cell of the crystal and @xmath183 denotes the various eigenmodes at @xmath108 .
physically equation means that _ local charge neutrality _ of the elementary cell is maintained in the long - wavelength limit @xmath184 , i.e. the mode induced ct is organized within the cell such that from the outside the cell looks electrically neutral .
equation puts a strong constraint on the possible ct in the cell and thus on the screening by cf s .
this restriction explains the small renormalization of the frequency of only 0.581 thz of @xmath104 in relation to @xmath12 where the renormalization due to cf s is 4.046 thz .
in @xcite the pronounced softening and the broadening of @xmath81 is attributed to a strong coupling between this phonon and an interlayer charge transport .
such an interpretation is supported by our calculations for @xmath12 where the cf s can be read off from and the ct is illustrated in in terms of the nonlocal part of the displacement induced rearrangement of the charge density according to @xmath185 the cf s @xmath186 in equation are obtained from equation and the form - factors from equation .
in contrast to the constraint for @xmath104 expressed by equation no such restriction must be satistfied in the metallic phase for @xmath12 . while in @xmath104
only an _ intracell _ ct is allowed summing up to zero , @xmath81 generates cf s of the same sign in the cell ( ) . this finally leads to cf s of alternating sign in consecutive ruo layers ( cells ) , i.e. an _ interlayer ( intercell ) _ ct is set up , see also .
the latter provides an effective screening mechanism of the coulomb interaction and explains the strong renormalization of @xmath81 as compared with @xmath104 . in the adiabatic approximation used in the calculations so far
the interlayer ct is instantaneous .
it has to be be replaced by a dynamic collective charge transfer in case a phonon - plasmon scenario would be realistic as for la@xmath0cuo@xmath1 in a nonadiabatic region nearby the @xmath3-axis .
the question if such a scenario is also likely for the more anisotropic sr@xmath0ruo@xmath1 is discussed in . and @xmath187 branch in la@xmath0cuo@xmath1 @xcite ( a ) and sr@xmath0ruo@xmath1 ( b ) . in ( b )
these branches have been obtained for three different values of the on - site repulsion u(ru4d ) of the ru4d orbitals and are characterized by different line types , u(ru4d)=0.619 dry ( @xmath90 ) , u(ru4d)=0.819 dry ( @xmath188 ) and u(ru4d)=1.005 dry ( -@xmath189-@xmath189- ) .
the experimental values are indicated as open squares @xmath190 @xcite . ]
@*7l & & @xmath191 & 38.252 & 26.603 & 17.871 & 50.875 & 33.764 & 22.048 + @xmath192 & 19.635 & 20.642 & 21.381 & 21.178 & 22.875 & 24.031 an important difference is found when considering the obsm in the cuprates and in sr@xmath0ruo@xmath1 , respectively .
the anomalous softening of the obsm phononanomalies , see ( a ) , in particular of the half breathing mode @xmath106 which is typical for la@xmath0cuo@xmath1 @xcite , and most probably generic for all the cuprate based htsc s @xcite , @xcite is strongly reduced in our calculation as compared with la@xmath0cuo@xmath1 . in the experiments
the softening is completely absent , see ( b ) .
our calculations show that there is a dominant microscopic reason for the vanishing of the obsm anomalies in sr@xmath0ruo@xmath1 , namely the magnitude of the on - site coulomb repulsion u(ru4d ) of the 4d orbitals of ruthenium .
the latter is obviously underestimated in our ab - initio calculation based on the ionic form - factor @xmath193 .
compared with the calculated value for the on - site repulsion u(cu3d)=1.005 dry in la@xmath0cuo@xmath1 which leads to good results for the frequencies of the obsm in this material , the calculated value for ruthenium , u(ru4d ) = 0.619 dry , seems not so reliable because of the more incomplete 4d shell of ru and the spherical average assumed in our computations for the orbital densities .
so , we increased the value of the ru4d on - site repulsion in two steps towards the calculated result for the cu3d orbital for comparison . as a consequence
the phonon anomalies completely vanish in these calculations in agreement with experiment as displayed in ( b ) .
simultaneously by increasing u(ru4d ) the cf s on the ru4d orbitals that are a measure for the nonlocal electron - phonon coupling are strongly decreased as can be seen in .
branches displaying a complex anticrossing behaviour .
the open squares @xmath190 are the experimental data @xcite . ] in @xcite and @xcite sharply peaked magnetic fluctuations at @xmath194 have been observed for sr@xmath0ruo@xmath1 .
these fluctuations have been related to dynamical nesting properties between the flat @xmath8 and @xmath9 sheets of the fs , compare with ( a ) .
braden et al .
@xcite have performed a search for kohn anomalies in the phonon spectrum related to this nesting structure .
promising candidates would be the high frequency @xmath187 modes which are coupled to the ru4d cf s . however
no signature for a kohn anomaly is present in the experiments as well as in the calculation in .
another possibility discussed in @xcite is the longitudinal acoustic @xmath187 branch where a small dip is observed just around @xmath195 , see . from our calculations
it seems very unlikely that this dip is a nesting effect because our computations of the phonon dispersion shows that this feature is well described by a complex anticrossing between the three lowest branches with @xmath187 symmetry , as shown in .
so we do not find in sr@xmath0ruo@xmath1 any evidence for epi driven by nesting of the fs .
we investigate for sr@xmath0ruo@xmath1 the possibility of a phonon - plasmon scenario along the @xmath3-axis and in a small region around this axis by performing calculations of the coupled electron - phonon dynamics in the nonadiabatic regime .
this is achieved approximatively by allowing the proper polarization part @xmath78 of the drf to depend on the frequency of the perturbation .
different from _
static screening _ in the adiabatic approximation this leads to _ dynamical screening _ of the coulomb interaction . a nonadiabatic treatment of the charge response and a related phonon - plasmon coupling has been shown recently to be a realistic option for la@xmath0cuo@xmath1 if the real anisotropy of the material is taken into account in the theory @xcite .
$ ] at the @xmath13 point in arbitrary units for sr@xmath0ruo@xmath1 in the m27bm2 ( a ) and for la@xmath0cuo@xmath1 in the m31bm @xcite ( b ) .
the real part is given by the full line ( @xmath90 ) and the imaginary part by the dotted line ( @xmath188 ) . ]
due to the much weaker @xmath2-dispersion of the electronic band - structure obtained in the computation for sr@xmath0ruo@xmath1 as compared with la@xmath0cuo@xmath1 we can expect much lower plasmon frequencies along the @xmath3-axis .
this conjecture can be shown to be true for example by calculating at the @xmath13 point the free - plasmon frequency according to equation and checking the result against la@xmath0cuo@xmath1 , see . from the zero crossing of the real and the imaginary part of @xmath196 $ ] we get for sr@xmath0ruo@xmath1 in the m27bm2 the very small value of @xmath197 mev ( 1.342 thz ) while for la@xmath0cuo@xmath1 we have @xmath198 mev ( 9.509 thz ) @xcite which is by a factor of seven larger .
this also expresses the much stronger anisotropy of sr@xmath0ruo@xmath1 in relation to the cuprates . in the analysis of the phonon - plasmon scenario
we have assumed the collisionless regime , i.e. the quasiparticles ( qp ) do not scatter each other and damping is only possible via intraband electron - hole decay and low - lying interband transitions .
however , in the cuprates and also in sr@xmath0ruo@xmath1 qp scattering due to electron - electron interaction and other degrees of freedom is certainly important .
this leads to damping and possibly overdamping of @xmath3-axis plasmons at sufficiently low frequencies and so the latter may cease to be well defined collective excitations of the fermi liquid . in order to make an estimate of the range of frequencies where @xmath3-axis plasmons should exist we examine the zero crossing of the real part of the well known dielectric function of an electron gas @xmath199 @xmath200 is the high - frequency dielectric constant , @xmath201 the scattering rate of the qp s along the @xmath3-axis . equating @xmath202 to zero we have the equation , @xmath203 using for @xmath10 the value given in @xcite in the normal state of la@xmath0cuo@xmath1 and the calculated value for the high - frequency dielectric constant along the @xmath3-axis @xmath204 @xcite we obtain @xmath205 for @xmath206 thz . for smaller plasmon frequencies we have @xmath207
, i.e. the plasmon has a diffusive pole and hence the @xmath3-axis plasmon is overdamped in the normal state .
it should be remarked that larger values for @xmath208 would increase the critical value for @xmath209 beyond that the plasmon could exist .
for example setting @xmath210 or @xmath211 we would have @xmath212 thz or 10.19 thz , respectively . assuming a similar @xmath3-axis scattering rate of the qp s for sr@xmath0ruo@xmath1 we regard @xmath213 thz as a reasonable limiting frequency beyond which plasmons most likely are not overdamped .
modes ( @xmath90 ) in the collisionless regime , the coupling modes at @xmath108 @xmath214(ferro , na ) and at @xmath215(na ) ) are shown as black dots ( @xmath216 ) .
( b ) @xmath217 modes ( @xmath90 ) along the @xmath218 direction from @xmath219 to @xmath220 .
the black dot ( @xmath216 ) at @xmath13 and @xmath221 are the @xmath12(na ) and approximately the @xmath222 mode in adiabatic approximation , respectively . in both figures
the broken line ( -@xmath189-@xmath189- ) denotes the free - plasmon branch and the dotted line ( @xmath188 ) the borderline for damping due to electron - hole decay . ]
thus , from our calculations of the plasmon along the @xmath3-axis ( @xmath110 direction ) displayed in ( a ) which are at very low frequencies around 1 thz we conclude that the latter are overdamped .
on the other hand at larger plasmon frequencies which result in case the wavevector @xmath46 is not strictly parallel to the @xmath3-axis , as shown for example in ( b ) , the plasmon can acquire enough energy to resist overdamping and may exist as a collective excitation .
in ( a ) we identify instead of the six @xmath6 branches of the adiabatic approximation ( compare with ( c ) ) an additional branch due to phonon - plasmon coupling .
the free - plasmon has frequencies around 1 thz and is nearly dispersionless along the @xmath110 direction .
this is in contrast with the situation in la@xmath0cuo@xmath1 @xcite where the free - plasmon has significant higher frequencies and increases from about 9.5 thz at the @xmath13 point to about 12.8 thz at @xmath108 .
thus , the plasmon is not overdamped along the @xmath3-axis in la@xmath0cuo@xmath1 while the phonon - plasmon scenario calculated in the collisionless regime for sr@xmath0ruo@xmath1 as displayed in ( a ) is not realistic . , @xmath2 ) plane with the directions @xmath223 , @xmath224 and @xmath225 .
@xmath226 $ ] .
na : nonadiabatic region ; ad : adiabatic region . ] analogous to the situation in la@xmath0cuo@xmath1 @xcite the calculated free - plasmon frequencies rapidly increase in proportion to the transverse component of the @xmath46 vector in a small nonadiabatic region of the charge response around the @xmath3-axis .
we have performed calculations along the @xmath227 and @xmath228 direction in this region , see , for different small @xmath72-values to study a possible phonon - plasmon scenario in sr@xmath0ruo@xmath1 .
we do not go into the details of these computations here but what is important concerning the question of overdamping is that for @xmath229 in case of the @xmath230 direction and for @xmath231 in the @xmath232 direction the free - plasmon frequency is larger than 7 thz , the estimated value for the phonon - plasmon scenario to exist . from the calculations we find that the longitudinal @xmath102(ferro , na ) mode at @xmath108 ( ( a ) ) is virtually unchanged at @xmath233 and so does exist around 19 thz .
the same holds true for @xmath81(na ) at the @xmath13 point being practically the same as @xmath234 for @xmath235 in the @xmath232 direction .
similar as for la@xmath0cuo@xmath1 @xcite with increasing @xmath72 @xmath102(ferro , na ) and @xmath236(na ) rapidly leave the spectrum to high frequencies . in case of @xmath236(na )
this mode behaviour can also be extracted from ( b ) where we have presented our calculated results of the dispersion of the coupled phonon - plasmon @xmath237 modes along the @xmath238 direction from @xmath239 to @xmath240 for @xmath241 .
the highest mode at @xmath13 is the @xmath12(na ) mode also seen in ( a ) as a black dot .
the broken line is the dispersion of the free - plasmon branch calculated from equation and the dotted line is the borderline for damping due to electron - hole decay investigated from @xmath242 for the bands crossing the fermi level . from this figure
the range of the region with a nonadiabatic charge response can be estimated .
it is characterized by the steep branch which ultimately converges to the frequency of the adiabatic @xmath12 mode ( black dot at @xmath221 ) .
similar as in la@xmath75cuo@xmath76 this nonadiabatic region is very small and can be estimated from the figure at about @xmath243 . unlike the case of la@xmath0cuo@xmath1 the phonon - plasmon dispersion in
( b ) should not exist in the full range of frequencies according to our estimate for overdamping below about 7 thz .
parallel to the investigation of the @xmath12 mode in la@xmath0cuo@xmath1 @xcite , the line - broadening of this mode observed by @xcite in sr@xmath0ruo@xmath1 can be understood from the calculated phonon - plasmon scenario .
experimentally there is a limited wave vector resolution in ins for the transverse direction perpendicular to the @xmath3-axis which for the experiments in la@xmath0cuo@xmath1 is on the average @xmath244 @xcite .
thus , the relevant frequency range sampled in the measurement for @xmath81 is over the steep branch beyond which the mode develops an @xmath12-like displacement pattern .
according to the calculations this sampling occurs for mode frequencies larger than 12.5 thz and leads to a corresponding broadening of the linewidth for @xmath81 observed in the experiment . because the region with a metallic adiabatic charge response outweights by a factor of roughly three the nonadiabatic region we can attribute to the @xmath12 mode an adiabatic frequency of about 15.3 thz .
the @xmath217 branch starting at @xmath12(na ) in ( b ) has not been observed in the experiments so far because a very high @xmath46-space solution transverse to the @xmath3-axis would be needed to resolve the mode dispersion in the small nonadiabatic sector .
in @xcite we have pointed out for la@xmath0cuo@xmath1 the relevance of a strongly coupling nonadiabatic @xmath12 mode .
so it is interesting to see if such a large epi also exists for @xmath81(na ) in sr@xmath0ruo@xmath1 . as a parameter for the strength of the epi in a certain mode @xmath55 we have used in the past the orbital averaged changes of the selfconsistent crystal potential @xmath245 , @xcite . in case of la@xmath0cuo@xmath1 we obtain in units of mev @xmath246 and @xmath247 which should be compared with the considerably reduced strength @xmath248 and @xmath249 in sr@xmath0ruo@xmath1 .
altogether , we conclude from our calculations for sr@xmath0ruo@xmath1 that nonadiabatic epi via phonon - plasmon coupling is not possible strictly along the @xmath3-axis but most likely away from this axis in a very small region at higher free - plasmon frequencies .
moreover , the strength is significantly smaller as in la@xmath0cuo@xmath1 .
we have argued above that the longitudinal ferroelectric @xmath102(ferro , na ) mode exists in the nonadiabatic sector around the @xmath108 point at about 19 thz .
this nonadiabatic result helps to understand why @xmath3-axis optical activity seen in the experiments @xcite is possible despite the fact that sr@xmath0ruo@xmath1 is in the metallic phase . in @xcite three @xmath3-polarized @xmath86 phonon structures at 6.04 thz , 10.88 thz and 14.51 thz
have been detected which agree well with the calculated transverse @xmath102 modes in the m27bm2 at 5.16 thz , 10.42 thz and 14.49 thz .
the by far largest oscillator strength is obtained for the mode at 10.88 thz and according to our calculation this is the ferroelectric mode @xmath102(ferro ) .
consistent with the large oscillator strength of this mode in the experiment our nonadiabatic result for the longitudinal ferroelectric mode @xmath250 ( ferro , na ) predicts a very large @xmath86 splitting of about 9 thz .
this splitting dominates the infrared response for polarization along the @xmath3-axis .
further we find in the calculations two other longitudinal @xmath86 modes at 12.379 thz and 6.964 thz which define corresponding splittings with the transverse modes above . from a theoretical point of view
it is important to remark that optical activity in the metallic phase can not be explained using the adiabatic approximation for a calculation of the phonon dispersion as is commonly done by applying static dft for the metal @xcite . in such calculations
there will be no lo - to or @xmath86 splittings being a measure for the oscillator strength because the transverse effective charges vanish if the metallic phase is treated in adiabatic approximation and consequently the induced dipole moments defining the oscillator strength in the dielectric function ( matrix ) vanish too .
thus , there will be no optical activity in a metal by the phonons treated in adiabatic approximation which underlies static dft .
we have shown that for a reliable description of the @xmath3-axis charge response in sr@xmath0ruo@xmath1 and in particular the phonon dynamics along this axis a typical lda - based model is too isotropic and must be modified to account for the much weaker interlayer coupling in the real material .
similar as in the cuprates studied earlier the large anisotropy in sr@xmath0ruo@xmath1 is considerably underestimated in dft - lda calculations of the electronic bs . from our investigations
we conclude that an accurate representation of the very fainty @xmath2-dispersion of the bs nevertheless is essential to understand the @xmath3-axis polarized @xmath6 modes . while the lda - based model ( 27bm ) is insufficient a modified much more anisotropic model ( m27bm2 ) is well suited . we have examined the multi - sheet fs of sr@xmath0ruo@xmath1 with the two models and found that the experimental results for the fs are not well described within the 27bm . on the other hand
, the m27bm2 shows a good agreement with the measured fs .
this model can be considered to represent the real anisotropy in sr@xmath0ruo@xmath1 sufficiently well .
we have also calculated the magnitude of the strongly enhanced anisotropy in the m27bm2 compared with the 27bm more globally in terms of some fermi surface parameters important for transport properties .
checking both models against each other we find an enhancement of about a factor 18 for the anisotropy ratio of the drude plasma energy tensor and of about a factor of 10 for the fermi velocity tensor .
while the calculations demonstrate that sr@xmath0ruo@xmath1 is a nearly two - dimensional fermi liquid significantly more anisotropic than la@xmath0cuo@xmath1 the remaining very weak three - dimensionality is crucial to achieve a solid representation of the @xmath3-axis charge response and of certain @xmath3-axis phonons .
we have calculated in detail the phonon dispersion in sr@xmath0ruo@xmath1 along the main symmetry directions in the bz and compared the result with the dispersion of the htsc la@xmath0cuo@xmath1 being structural isomorphic .
a good overall agreement of the calculated results is found for sr@xmath0ruo@xmath1 with the ins data . comparing the results of the calculated phonon dynamics in the rim with a model allowing additional df s and
finally the full model including df s and cf s the renormalization of certain modes which are strongly coupled via nonlocal epi effects of df- and cf - type has been analyzed in detail . in this context
an important point of the comparison with la@xmath0cuo@xmath1 is that the anomalous softening of the high - frequency oxygen bond - stretching modes , being generic for the cuprate based htsc s , is strongly reduced or completely absent in sr@xmath0ruo@xmath1 depending primarily on the magnitude of the on - site coulomb repulsion of the ru4d orbitals .
we have also investigated the possibility of a kohn anomaly in the @xmath187 modes of sr@xmath0ruo@xmath1 as discussed in the literature . in our calculations
we do not find any evidence for such an anomaly driven by nesting of the fs .
instead , the dip in the lowest @xmath187 branch seen in the experiments is well explained in our computations by an anticrossing effect of the three lowest @xmath187 branches .
finally , we have examined the question of a possible phonon - plasmon scenario in a small region of nonadiabatic charge response around the @xmath3-axis which has been shown to be a realistic option in la@xmath0cuo@xmath1 .
due to the much weaker electronic @xmath2-dispersion obtained for sr@xmath0ruo@xmath1 the calculated free - plasmon frequencies along the @xmath3-axis are about a factor of eight smaller in the collisionless regime than in la@xmath0cuo@xmath1 .
we have argued that damping generated by the interactions between the qp s as well as interband transitions which are at much lower energy scale than in la@xmath0cuo@xmath1 leads to an overdamping of the plasmon strictly along the @xmath3-axis in contrast to la@xmath0cuo@xmath1 .
however , a coupled phonon - plasmon scenario becomes likely also in sr@xmath0ruo@xmath1 at higher free - plasmon frequencies .
this occurs in case the wavevector is not strictly parallel to the @xmath3-axis but has a small transverse component .
we find that the strength of this nonlocal nonadiabatic coupling is significantly weaker than in la@xmath0cuo@xmath1 .
ultimately we have explained by our calculations of the nonadiabatic charge response the linewidth of the apex oxygen breathing mode at the @xmath13 point and why @xmath3-axis optical activity as seen in the experiments is possible despite the fact that sr@xmath0ruo@xmath1 is in the metallic phase .
10 maeno y et al . 1994
* 372 * 532 bergemann c , mackenzie ap , julian sr , forsythe d and ohmichi e 2003 adv . phys . * 52 * 639 ho af and schofield aj 2005 phys . rev .
b * 71 * 045101 mackenzie ap , maeno y 2003 rev .
* 75 * 657 mao zq et al .
2003 phys .
b * 63 * 144514 falter c 2005 phys .
status solidi b * 242 * 78 bauer t and falter c 2008 phys .
b * 77 * 144503 falter c , bauer t , and schnetgke f 2006 phys . rev .
b * 73 * 224502 pintschovius l 2005 phys .
status solidi b * 242 * 30 falter c , klenner m and ludwig w 1993 phys .
b * 47 * 5390 braden m , reichardt w , sidis y and maeno y 2007 phys . rev .
* 76 * 014505 bauer t and falter c 2009 cond - mat/0808.2765 falter c , klenner m , hoffmann ga and schnetgke f 1999 phys . rev .
b * 60 * 12051 falter c 1988 phys .
* 164 * 1 falter c , klenner m and hoffmann ga 1995 phys .
b * 52 * 3702 falter c and schnetgke f 2002 phys . rev .
b * 65 * 054510 perdew jp and zunger a 1981 phys . rev .
b * 23 * 5048 krakauer h , pickett we and cohen re 1998 j. supercond .
* 1 * 11 savrasov sy and andersen ok 1996 phys .
. lett . * 77 * 4430 wang cz , yu r and krakauer h 1999 phys .
b * 59 * 9278 bohnen kp , heid r and krauss m 2003 europhys . lett .
* 64 * 104 giustino f , cohen ml and louie sg 2008 nature * 452 * 975 mazin ii , papaconstantopoulos da and singh dj 2000 phys .
b * 61 * 5223 shen km et al .
2007 phys .
lett . * 99 * 187001 chmaissem o et al .
1998 phys .
b * 57 * 5007 moore rg et al .
2008 phys .
lett . * 100 * 066102 uruma m et al .
2008 cond - mat/0711.2160 ( unpublished ) graf j , et al .
2008 phys .
lett . * 100 * 227002 sidis j et al .
1999 phys .
* 83 * 3320 braden m et al .
2002 phys .
b * 66 * 064522 tamasaku k , nakamura y and uchida s 1992 phys . rev . lett . * 92 * 1455 pintschovius l , private communication katsufuji t , kasai m and tokura y 1996 phys .
lett . * 76 * 126 | the phonon dynamics of the low - temperature superconductor sr@xmath0ruo@xmath1 is calculated quantitatively in linear response theory and compared with the structurally isomorphic high - temperature superconductor la@xmath0cuo@xmath1 .
our calculation corrects for a typical deficit of lda - based calculations which always predict a too large electronic @xmath2-dispersion insufficient to describe the @xmath3-axis response in the real materials . with a more realistic computation of the electronic band structure the frequency and wavevector dependent irreducible polarization part of the density response function
is determined and used for adiabatic and nonadiabatic phonon calculations .
our analysis for sr@xmath0ruo@xmath1 reveals important differences from the lattice dynamics of @xmath4- and @xmath5-doped cuprates .
consistent with experimental evidence from inelastic neutron scattering the anomalous doping related softening of the strongly coupling high - frequency oxygen bond - stretching modes ( obsm ) which is generic for the cuprate superconductors is largely suppressed or completely absent , respectively , depending on the actual value of the on - site coulomb repulsion of the ru4d orbitals .
also the presence of a characteristic @xmath6-mode with a very steep dispersion coupling strongly with the electrons is missing in sr@xmath0ruo@xmath1 .
moreover , we evaluate the possibility of a phonon - plasmon scenario for sr@xmath0ruo@xmath1 which has been shown recently to be realistic for la@xmath0cuo@xmath1 .
in contrast to la@xmath0cuo@xmath1 in sr@xmath0ruo@xmath1 the very low lying plasmons are overdamped along the @xmath3-axis . |
chirality is a geometric property of a three - dimensional body not to coincide with its reflection in a mirror in any shifts and turns .
such a property , for example , belongs to a human hand or a spring . the term `` chirality '' was proposed by lord kelvin in 1873 to explain some special properties of molecules @xcite .
the chirality plays an important role in biology and pharmacy , because many complex organic compounds ( amino acids , proteins , and sugars ) have chiral properties .
for this reason , a body can react differently with different enantiomers of the same substance .
for example , the same drug , depending on what type of molecules it contains , may have a different taste and smell , or effect differently . in physics
, there is a considerable interest to chiral media @xcite , in which there is a difference between the distributions of left- and right - handed circularly polarized electromagnetic waves .
the simplest example of such an optically active medium is an aqueous solution of sugar . at the present time
, the study of the chiral properties is again under special attention because of the possibility to create metamaterials based on chiral objects @xcite .
the spontaneous emission of atoms and molecules in the vicinity of material bodies is studied in a large number of works .
the influence of dielectric microspheres on the spontaneous emission of atoms is considered in @xcite , spontaneous emission of an atom located near a microsphere made of material with negative refraction was considered in @xcite for first time .
optical properties of chiral spherical particles have also been studied quite extensively . at the present time , the scattering of plane electromagnetic waves on homogeneous @xcite and inhomogeneous @xcite chiral microspheres , including multilayered chiral microspheres @xcite , is already considered .
the scattering of a hermitian laser beam on a chiral microsphere is studied in @xcite , the scattering of a plane electromagnetic wave by a chiral sphere located in a chiral medium is considered in @xcite .
the radiation pressure force acting on a chiral spherical particle placed in circularly polarized wave is examined in @xcite .
finally , the radiation of a point dipole source located inside a chiral spherical particle was regarded in @xcite . at the same time , as far as we know , there is no investigation devoted to the influence of chiral spherical particles on the radiation of chiral ( optically active ) molecules .
however , such processes become more and more important because they always appear when different biomolecules and drug molecules are investigated and modified by optical methods .
so , in this paper we will present the detailed investigation of this problem the goal of the present work is to develop an approach that allows describing quantitatively the influence of a chiral spherical particle made of an arbitrary metamaterial on the spontaneous emission of a chiral molecule .
all results will be obtained in the framework of nonrelativistic quantum electrodynamics , in the assumption that the radiative linewidth of the molecule is much smaller than the frequency of radiation .
the last assumption allows us to use the perturbation theory .
the rest part of the paper is organized as follows . in the section [ sect2 ] ,
the canonical quantization of the electromagnetic field in the presence of a chiral spherical particle with arbitrary permittivity and permeability is performed . in the section [ sect3 ]
, we obtain general expressions for the spontaneous emission radiative decay rate of a chiral molecule located near a chiral sphere . in the section [ sect4 ] , the graphical illustration of the results obtained and their discussion
are presented .
the geometry of the problem is shown in fig .
to solve the problem of quantization of the electromagnetic field in the presence of a chiral spherical particle , let us consider a spherical cavity of an infinite radius @xmath0 with a perfectly conducting wall and with the a chiral particle located in its center . for simplicity ,
we assume that the nanoparticle is surrounded by vacuum ( see fig .
[ fig1 ] ) .
the usual procedure of quantization of the electromagnetic field in spherical geometry @xcite is inapplicable in the case of chiral particles , and below we will develop a new approach that is valid in this case . in the description of classical fields in the presence of chiral spherical particles , we will use the method proposed in @xcite . at the same time , to describe the chiral medium , we use the constitutive equations in the fedorov s form @xcite : @xmath1 where @xmath2 , @xmath3 and @xmath4 , @xmath5 are the inductions and the strengths of electric and magnetic fields , correspondingly , @xmath6 , @xmath7 are the dielectric permittivity and the magnetic permeability of the chiral medium , and @xmath8 is the dimensional parameter of chirality . substituting equations ( [ eq1 ] ) into maxwell s equations , one can obtain @xmath9 where @xmath10 is the wave number in vacuum , and @xmath11 is the dimensionless parameter of chirality . in eq .
( [ eq2 ] ) and everywhere further , the time dependence @xmath12 is assumed .
if we rewrite the system ( [ eq2 ] ) in the matrix form : @xmath13 then , with the help of the bohren s transformation for the electric and magnetic fields @xcite : @xmath14 the matrix @xmath15 can be diagonalized : @xmath16 where @xmath17 are the wave numbers of the left- ( @xmath18 ) and right - polarized ( @xmath19 ) waves .
the components of the transformed field ( [ eq4 ] ) satisfy the equations @xcite : @xmath20 thus , the fundamental solutions of the maxwell s equations in a chiral medium are the left - polarized and right - polarized waves with different wave numbers . in the spherical geometry under consideration , in the case of an arbitrary electromagnetic field , the expressions for @xmath21 and @xmath22 can be written down as follows ( we will omit a multiplier @xmath23 here and below ) : @xmath24 where @xmath25 and @xmath26 are the coefficients of expansions .
these coefficients can be found with the help of additional conditions .
explicit expressions for spherical vector harmonics @xmath27 and @xmath28 ( where @xmath29 , @xmath18 ) are given in the appendix . from eq .
( [ eq8 ] ) , it follows that @xmath21 and @xmath30 are expressed in terms of fixed combinations of vector spherical harmonics @xmath31 and @xmath32 , which always have a nonzero component along the radius ( see the appendix ) and therefore never can be reduced to the usual te and tm fields . in accordance with the general rules of quantization of the electromagnetic field
, its expansion in the complete system of eigenfunctions of the classical problem ( standing spherical waves ) can be represented as follows ( the indices are omitted ) :
@xmath33 where @xmath34 and @xmath35 are the operators of annihilation and creation of a photon in @xmath36-th mode respectively , with usual rules of the commutation , @xmath37 and @xmath38 are the modes of the electric and the magnetic field , the vector index @xmath39 denotes a chosen set of quantum numbers : the orbital ( @xmath40 ) , the azimuthal ( @xmath41 ) , and the radial ( @xmath42 ) , the asterisk denotes the operation of the complex conjugation . in accordance with eqs .
( [ eq4 ] ) and ( [ eq8 ] ) , the fields inside the chiral sphere can be written as @xmath43 outside of the chiral particle , all waves have the same propagation velocity and can be represented as @xmath44 where @xmath45 , @xmath46 and @xmath47 , @xmath48 are some coefficients , and expressions for the spherical vector harmonics @xmath49 and @xmath50 ( where @xmath51 , 2 ) are given in the appendix . to find the coefficients in eq .
( [ eq11 ] ) , one should use the condition of continuity of tangential components of the electric and magnetic fields on the surface of a chiral spherical particle having the radius @xmath52 . since the molecule is situated outside the particle
, we do not need to know explicit values of the coefficients @xmath25 and @xmath53 ( see eq .
( [ eq8 ] ) ) . for the coefficients that describe fields outside of the sphere
, one can obtain the following relations from the boundary conditions : @xmath54 in which @xmath55 in eq .
( [ eq13 ] ) , we have used the abbreviations ( @xmath56 , @xmath19 ) : @xmath57 where @xmath58 , @xmath59 and @xmath60 are the riccati - bessel functions ( see the appendix ) , and the prime near the function means the derivative with respect to its argument . to study the interaction between a two - level molecule and the continuum of electromagnetic modes modified by the presence of a chiral spherical particle , it is necessary to know the density of final states . to find it
, one should use the condition of disappearance of tangential components of electric field on the inner surface of the cavity . as a result
we have @xmath61 where @xmath62 is the riccati - bessel function ( see the appendix ) , and the prime near the function means its derivative . substituting eq .
( [ eq12 ] ) into eq .
( [ eq15 ] ) and using asymptotic expressions for the hankel function at @xmath63 @xcite , we obtain @xmath64 the system ( [ eq16 ] ) has nontrivial solutions only if its determinant equals to zero . as a result
, we have a quadratic equation , which determines two different relationships between the coefficients @xmath65 and @xmath66 @xmath67 the relations ( [ eq17 ] ) define two types of electromagnetic modes ( photons ) occurring in the presence of a chiral spherical particle , which are analogous to tm and te modes in the case of a nonchiral spherical particle .
the upper sign ( @xmath68 ) in eq .
( [ eq17 ] ) corresponds to modes which we will call @xmath69-type photons , while another sign ( @xmath70 ) gives us @xmath71-type photons .
if the chirality parameter tends to zero ( @xmath72 , then , as it follows from eqs .
( [ eq13 ] ) and ( [ eq14 ] ) , the right side of eq .
( [ eq17 ] ) tends to zero or to infinity , which leads to a decoupling between te and tm modes ( vector spherical harmonics @xmath73 and @xmath74 in eq .
( [ eq11 ] ) ) , and as a result leads to the appearance of usual tm or te polarized photons .
thus , in the case of a chiral spherical particle we have two types of polarization ( @xmath69 and @xmath71 ) , which are different from the tm and te polarizations . substituting eq .
( [ eq17 ] ) into eq .
( [ eq16 ] ) , we find the next asymptotic expressions ( @xmath0 ) : @xmath75 where @xmath76 is the mode eigenfrequency , and @xmath77 is the speed of light in vacuum , and @xmath78 from eq .
( [ eq18 ] ) , it follows immediately that the density of final states for both @xmath69-type and @xmath71-type photons has the same value : @xmath79 and does not dependent on the presence of a spherical particle of a final volume in the cavity , which is consistent with the courant s theorem @xcite . for the normalization of modes
, we will assume that there is only one photon in the quantization volume .
as a result , we obtain the condition : @xmath80 where @xmath81 is the kronecker s delta , and the inductions @xmath82 and @xmath83 of photonic modes can be expressed via the operators of field strengths @xmath84 and @xmath85 with the help of the constitutive equations ( [ eq1 ] ) . to calculate the integral in eq .
( [ eq21 ] ) , we note that the main contribution to it comes from the region @xmath86 outside the spherical particle , where there is no chirality and where it is possible to use asymptotic expressions for the hankel functions . as a result , we obtain ( @xmath87 , 2 , 3 , @xmath88 ; @xmath89 , @xmath90 , @xmath91 , @xmath88 , @xmath92 ) : @xmath93 thus , expression ( [ eq11 ] ) , ( [ eq12 ] ) , and ( [ eq22 ] ) , together with the expression for the density of the final states ( [ eq20 ] ) , fully describe the quantized electromagnetic field in the presence of a chiral spherical particle , and allow to describe the interaction of this field with arbitrary atoms and molecules . in the next section , we will apply these expressions to find the radiative decay rate of the spontaneous emission of a chiral molecule .
the radiative decay rate of the spontaneous emission ( the radiative linewidth @xmath94 ) of a chiral molecule located near a chiral spherical particle can be found by making use of the fermi s `` golden rule '' @xcite : @xmath95 where summation is performed over all possible final states , which include both types of photons ( @xmath69 and @xmath71 , see eq .
( [ eq17 ] ) ) , @xmath96 is the density of the final states ( [ eq20 ] ) , and @xmath97 is the interaction hamiltonian of a chiral molecules with the electromagnetic field . in this work , we assume that the hamiltonian is hermitian , i.e. there are no losses in the material of a particle .
if there are losses , there is a possibility of nonradiative transition from the excited state to the ground state .
the expression for the nonradiative part of the spontaneous decay rate can be obtained within the framework of the classical approach @xcite or by using the nonstandard quantization scheme @xcite . in further calculations , for brevity we will consider only one channel of the decay of the initial state of the molecule ( transition @xmath98 ) , i.e. the two - level molecule . to take into account the possibility of transition of the molecule into several states ,
it is enough to sum the partial linewidths .
the hamiltonian of interaction of a chiral molecule with the electromagnetic field in eq .
( [ eq23 ] ) can be written down in the next form @xcite : @xmath99 where @xmath100 is the radius vector of the molecule position , @xmath101 and @xmath102 are operators of the electric and the magnetic dipole moments of the molecule , and field operators @xmath103 and @xmath104 are defined in eq .
( [ eq9 ] ) .
we suppose that the initial state corresponds to the vacuum state of the field and to the molecule in the excited state , @xmath105 , and that the final field state correspond the single photon ( of @xmath69-type or @xmath71-type ) and the molecule in the ground state @xmath106 . besides
, for definiteness we will consider a spiral model of a molecule in which electrons are constrained to move along a helical path .
substituting eq .
( [ eq24 ] ) into eq .
( [ eq23 ] ) , we obtain the following expression for the total radiative decay rate of spontaneous emission of the molecule : @xmath107 where , for example , in the case of photons of @xmath69-type @xmath108 where @xmath109 and @xmath110 are dipole moments of the considered transition of the molecule on the frequency @xmath111 ( see eq .
( [ eq18 ] ) ) .
the chosen phase difference between the electric and magnetic dipoles is due to the fact that the magnetic moment operator is purely imaginary .
this definition also agrees with the adopted spiral model of a molecule . for more complicated chiral molecules
, there can be a different phase relation between electric and magnetic dipole matrix elements .
let us , for clarity , call the molecules with parallel @xmath112 and @xmath113 the `` right '' molecules , while the molecules with antiparallel @xmath112 and @xmath113 will be referred to as the `` left '' molecules .
explicit expressions for the component of eq .
( [ eq25 ] ) for the mode of photons of @xmath71-type one can obtain from eq .
( [ eq26 ] ) by replacing the index @xmath114 .
let us assume for definiteness that the molecule is located on the z - axis of a cartesian coordinate system at the point @xmath115 . in this case , only components with @xmath89 , @xmath90 are nonzero , and the explicit expressions for eq .
( [ eq25 ] ) will have the following form : @xmath116 where @xmath117 @xmath118 and @xmath119 in the case of photons mode of @xmath71-type , one can obtain @xmath120 where @xmath121 @xmath122 and @xmath123 in eqs .
( [ eq28])-([eq30 ] ) , ( [ eq32])-([eq34 ] ) , we have used the following notations : @xmath124 in the case when the chirality parameter of a spherical particle is equal to zero , eqs .
( [ eq28])-([eq30 ] ) , ( [ eq32])-([eq34 ] ) are simplified and have the following form : @xmath125 @xmath126 @xmath127 and @xmath128 @xmath129 @xmath130 where @xmath131 eqs .
( [ eq36])-([eq41 ] ) correspond to components of the radiative decay rate of spontaneous emission of a chiral molecule located near the dielectric sphere .
if we put @xmath132 or @xmath133 in these expressions , it will be possible to find known relations for the radiative decay rate of spontaneous emission of an atom located near a dielectric spherical particle , and which has nonzero electric or magnetic dipole moment , respectively @xcite .
note , that despite of the fact , that this derivation was performed in assumption of absence of losses in the material of the particle , the expressions obtained will also exactly describe radiation losses for nanoparticles made of real materials , i.e. for complex @xmath6 and @xmath7 .
it follows from the fact that classical and quantum calculations give the same analytical expressions for the radiative decay rate of spontaneous emission through the green s function of a classical problem @xcite .
since the classical approach is valid in the case of both absorbing and nonabsorbing bodies , this means that for a generalization of the expressions obtained for the case of a real ( absorbing ) material of a spherical particle , it is only necessary to substitute appropriate complex values of dielectric permittivity and magnetic permeability into obtained expressions . to calculate the relative radiative decay rate of spontaneous emission of chiral molecules located near a chiral spherical particle , one should normalize the expression ( [ eq25 ] ) to the value of the radiative decay rate of spontaneous emission of the molecule in the absence of the particle .
for this purpose , we use eqs .
( [ eq36])-([eq41 ] ) where we put @xmath134 . by substituting the expressions obtained in such a way into eqs .
( [ eq27 ] ) , ( [ eq31 ] ) and ( [ eq25 ] ) , and performing summation , it is possible to find the radiative decay rate of spontaneous emission of a chiral molecules in free space : @xmath135 this expression , of course , coincides with the sum of the known expressions for the linewidths of electric and magnetic dipoles @xcite , because in the absence of chirality the interference of radiation does not occur .
cumbersome formulas obtained in the previous section can not fully reveal the features of the spontaneous decay of a chiral molecule in the system under consideration . in this section
, we will explore the expressions found in the typical special cases . in a very important case of a chiral nanoparticle ,
i.e. a particle having a size much smaller than the wavelength of the radiation , the main contribution to eqs .
( [ eq28])-([eq30 ] ) , ( [ eq32])-([eq34 ] ) is from term with @xmath40 = 1 . expanding the coefficients of eq .
( [ eq13 ] ) in a series of @xmath136 and keeping only the principal terms , we obtain @xmath137 the condition that the denominator of eq .
( [ eq44 ] ) will be a zero @xmath138 corresponds to the condition of a chiral - plasmon resonance occurring in a spherical nanoparticle .
the roots of eq .
( [ eq45 ] ) determine values of the constants @xmath6 and @xmath7 , that correspond to a chiral plasmon .
it should be noted , that the resonant values of @xmath6 and @xmath7 are not independent , and their values differ from the corresponding values in the case of plasmon resonance in a nonchiral particle , where ( [ eq45 ] ) is split into two independent equations , @xmath139 and @xmath140 @xcite . to find an expression for the spontaneous emission radiative decay rate in the case of a nanoparticle , one should expand @xmath87 terms in eqs .
( [ eq28])-([eq30 ] ) , ( [ eq32])-([eq34 ] ) into a series of @xmath136 and keep only the principal terms . however , it is a too long procedure , and here we will apply a different approach used in @xcite .
if the nanoparticle and the molecule are placed close enough , then their total radiation will be determined by their total electric and magnetic dipole moments : @xmath141 where @xmath142 and @xmath143 are the electric and the magnetic dipole moments induced in the nanoparticle while @xmath112 and @xmath144 are the oscillating dipole moments of the molecule . explicitly , these values can be written as follows @xcite : @xmath145 where the polarizability of a chiral spherical nanoparticle in a homogeneous field is @xmath146 the fields in eq . ( [ eq47 ] )
have the form : @xmath147 where @xmath148 is the unit vector in the direction from the center of the particle to the observation point . in practice
, the orientation of the molecule can be arbitrary with respect to the nanoparticle surface , therefore to get an effective radiative decay rate of spontaneous emission one should average eq .
( [ eq46 ] ) on the orientations of molecules , or , equivalently , on the unit vector @xmath148 . as a result
we obtain @xmath149 where @xmath150 is defined by the relation @xmath151 . as a rule ,
the magnetic dipole moment of the molecule is much smaller than the electric dipole moment @xmath152 .
the chirality parameter , even in hypothetical metamaterials , is also small ( @xmath153 ) .
this fact determines that the second term in eq .
( [ eq50 ] ) , corresponding to the induced electric dipole moment , is usually greater than the term corresponding to the induced magnetic dipole moment .
thus , the effective interference between the electric and magnetic fields is possible only if the following two conditions take place : _ 1 . in the system under consideration , a chiral - plasmon resonance must be present , i.e. the condition ( [ eq45 ] ) must be satisfied . under this condition , the contribution of a magnetic radiation increases .
_ _ 2 . the electric dipole moment induced in the nanoparticle should be zero , i.e. the next condition must be satisfied ( see eq .
( [ eq50 ] ) ) _
@xmath154 the solution of the system of eqs .
( [ eq45 ] ) and ( [ eq51 ] ) determines the values of dielectric permittivity and magnetic permeability of the nanoparticle , which correspond to minimal values of the radiative decay rate of a chiral molecule .
it means that the interference between the electric dipole and magnetic dipole radiation becomes maximal and destructive if @xmath155 from eq .
( [ eq52 ] ) , it follows that by changing the sign in @xmath156 , i.e. when the chirality of the molecule is changed , the `` resonant '' magnetic permeability @xmath157 varies only slightly ( @xmath158 , @xmath159 ) and approximately equals to @xmath160 . on the other hand
, the `` resonant '' dielectric permittivity @xmath161 may have different signs for molecules with different chirality .
this means that both nanoparticles with simultaneously negative @xmath162 and @xmath7 ( double negative - dng - metamaterials ) , and nanoparticles with negative @xmath7 and positive @xmath6 ( @xmath7 negative - mng - metamaterials or magnetic plasma ) are suitable for the effective control of radiation of chiral molecules . from the viewpoint of practical implementation of such nanoparticles ,
the most suitable are nanoparticles with a negative @xmath7 and positive @xmath6 , because they can be implemented by using a well developed technology of split rings resonators @xcite . on the other ,
the chiral dng - metamaterials can be also synthesized @xcite .
the above expressions give a good description of properties of arbitrary chiral molecules radiation of near nanoparticles , i.e. , when the retardation can be ignored . in the case
when retardation effects are significant , it is necessary to use the full expression for the decay rates given in the previous section . in any case , the process of spontaneous decay of a chiral molecule located near a chiral spherical particle is very complex and below for clarity we will give graphical illustrations of some possible regimes of interaction of a chiral molecule with nanoparticles of different composition but neglecting losses in them .
the figure [ fig2 ] shows the distribution of absolute values of the coefficient @xmath163 ( see eq .
( [ eq35 ] ) ) depending on the dielectric permittivity and the magnetic permeability of the material of nonchiral ( fig .
[ fig2](a ) ) and chiral ( fig .
[ fig2](b ) ) spherical particles for different values of @xmath164 .
large values of this coefficient correspond to a resonance .
this figure clearly shows that there is a significant difference between these cases , as reflected in the fact that nonzero chirality of the particle leads both to an increase in the number of observed modes ( whispering gallery ( wg ) modes in dielectric `` right - handed '' ( rh ) material and in `` left - handed '' ( lh ) material , lh surfaces modes @xcite ) , and to the change of their structure as compared with a nonchiral particle .
this fact is due to the simultaneous presence of both left- and right - polarized waves in a chiral particle , which in turn leads to excitation of resonances analogous to tm and te resonances of nonchiral sphere , simultaneously .
the figure [ fig2 ] is very important because it allows one to qualitative understand the structure of the radiative decay rates of spontaneous emission of a chiral molecule placed near the different nanoparticles .
it is very interesting that even for a dielectric sphere with a small admixture of chirality , strong and unexpected effects occur . in fig .
[ fig3 ] , the radiative decay rate of spontaneous emission of a chiral molecule placed near a chiral spherical particle with positive values of permittivity and permeability is shown as a function of the size parameter . as it is seen in this figure
, the presence of chirality in a spherical particle leads to a shift of the maxima of the spontaneous emission radiative decay rate in comparison with the case of a nonchiral particle . indeed , as it is known , a condition that allows to estimate the position of the resonance is given by @xmath165 , where @xmath40 is the orbital quantum number .
in the case of chiral spherical particles , the left- and the right - polarized waves exist simultaneously , and @xmath166 if @xmath167 .
hence the above condition should be changed to @xmath168 . as a result
, this leads to smaller resonant values of @xmath164 than for a nonchiral particle ( see fig .
[ fig2 ] ) .
even more interesting feature of ordinary dielectric particles with a small admixture of chirality is a substantial increase in the quality factor of whispering gallery modes .
this figure clearly shows that the corresponding linewidth can be increased by the factor 9 or even more .
note also that in fig .
[ fig3 ] , only a slight difference between the radiative decay rates of spontaneous emission corresponding to the different orientations of the transition magnetic dipole moment of a chiral molecule at a fixed electric dipole moment is observed .
however , one should expect a growth of this difference in the case of a gradual increasing of absolute values of the magnetic dipole momentum .
figure [ fig4 ] shows the radiative decay rate of spontaneous emission of a chiral molecule located near chiral spherical nanoparticles ( @xmath169 ) as a function of the chirality @xmath170 . as it is clearly seen in this figure , changing the chirality of the sphere has the greatest impact on the rate of spontaneous decay of the molecules near dielectric and left - handed spherical particles , in which high - q modes can be excited .
when the chirality parameter approaches to the critical value @xmath171 , then the number of oscillations in the radiative decay rate increases rapidly , and @xmath172 ( see eq .
( [ eq6 ] ) ) . in the case of metallic particles , the dependence of the radiative decay rate of spontaneous emission on the parameter @xmath173 , which is shown in fig .
[ fig4 ] , is weak , due to imaginary values of @xmath174 , and due to the absence of propagating waves .
chiral molecules play an especially important role in biology and pharmacy .
therefore , it is extremely important to arrange the separation of `` right '' and `` left '' enantiomers of molecules in the racemic mixtures .
this can be done in various ways , for example by using radiation pressure forces in the electromagnetic field of left- or right - handed circularly polarized electromagnetic waves @xcite , or by using spiral optical beams @xcite .
in addition , such selection can be efficiently performed with the help of chiral nanoparticles @xcite .
figure [ fig5 ] shows the radiative decay rate of spontaneous emission of a chiral molecule located near the lh chiral spherical nanoparticles as a function of the dielectric constant for a fixed value of magnetic permeability . as it well seen in the figure , for molecules that differ only in the orientation of the magnetic dipole moment of the transition ( the `` left '' molecule with @xmath175 , and the `` right '' one with @xmath176 ) , the radiative decay rate takes significantly different values , and thus , under such conditions it is possible to control effectively the radiation properties of enantiomers .
it is important to note that both the results of calculations obtained in the framework of the full qed theory ( present paper ) , and in the framework of a simplified quasistatic calculations presented at the beginning of this section ( see also @xcite ) yield almost identical results , confirming the correctness of both approaches .
the figure [ fig6 ] shows the ratio of the effective radiative decay rate of spontaneous emission of the `` left '' molecule and the radiative decay rate of the `` right '' molecule and vice versa . from this figure
, it follows that if the condition ( [ eq52 ] ) is satisfied , then the decay rates of `` left '' and `` right '' molecules differ by the factor 15 or 60 or even more , depending on the chirality of the molecule considered as a reference one . in other words , nanoparticles with the parameter given by eq .
( [ eq52 ] ) will enhance the radiation of the `` right '' molecules and slow down the radiation of the `` left '' molecules and vice versa .
let us stress that `` left - handedness '' of a chiral sphere or its negative @xmath7 are of crucial importance for such discrimination .
possible applications of the effect of discrimination of the radiation one can find in @xcite .
thus , in the present work the analytical expressions for the radiative decay rate of spontaneous emission of a chiral molecule located near an arbitrary spherical particle with chiral properties were obtained and investigated within full qed theory . using this approach , the spontaneous emission decay rates of chiral molecule placed near a spherical particle made of different materials ( dielectrics , metals , `` left - handed '' metamaterials etc . )
were investigated in details .
it is shown that nonzero chirality of a spherical particle leads to an increase in the number of excited modes in comparison with nonchiral one due to coupling between left- and right - polarized waves .
quasistatic expressions for the radiative decay rate of spontaneous emission of chiral molecule located near a chiral nanosphere were obtained and investigated .
these expressions are in good agreement with the exact qed results and allow one to estimate the parameters of nanoparticles when the radiation of `` right '' or `` left '' molecules is suppressed .
it was found that for the suppression of the radiation of the `` left '' molecules the sphere should be made of mng metamaterials , while for the suppression of radiation the `` right '' molecules the dng materials should be used .
the results obtained can be used to calculate the radiative decay rate of spontaneous emission of chiral molecules in the vicinity of the chiral spherical particles , for interpretation of experimental data on interaction of chiral molecules and particles , and for detection and selection of chiral molecules with the help of chiral particles .
spherical vector harmonics that describe the electromagnetic field inside a chiral spherical particle have the form : where @xmath179 , @xmath180 and @xmath181 are spherical coordinates , @xmath182 , @xmath183 , @xmath184 are unit vectors of the spherical coordinate system , @xmath185 is the riccati - bessel function @xcite , where @xmath186 is the bessel function @xcite , the prime near the function means the derivative with respect to its argument , @xmath187 is the associated legendre function @xcite , and the index @xmath56 , @xmath19 . where @xmath190 is the riccati - bessel function @xcite , where @xmath191 is the hankel function of the first ( @xmath51 ) and second ( @xmath192 ) kind , respectively , and the prime near the function means the derivative with respect to its argument .
( color online ) the dependence of the absolute values of the coefficient @xmath163 on the permittivity ( @xmath193 and the permeability ( @xmath194 of chiral spherical particle for different values of @xmath164 .
( a ) the chirality parameter @xmath195 .
( b ) the chirality parameter @xmath196 .
the particle is placed in vacuum.,width=604 ] ( color online ) radiative decay rate of the spontaneous emission of a chiral molecule located in close vicinity to the surface of a chiral dielectric spherical particle ( @xmath197 ) with @xmath198 and @xmath199 as a function of its size , @xmath164 .
( a ) the transition electric dipole moment of the molecule is oriented along the @xmath200-axis ( @xmath201 ) , i.e. it is tangential to the surface of the particle .
( b ) the transition electric dipole moment of the molecule is oriented along the @xmath202-axis ( @xmath203 ) , i.e. it is normal to the surface of the particle .
the solid line corresponds to the transition magnetic dipole moment of a molecule oriented along the x - axis , and the dashed line corresponds to the transition magnetic dipole moment of a molecule oriented along the @xmath202-axis .
the value of the transition magnetic dipole moment @xmath204 .
the particle is placed in vacuum.,width=340 ] ( color online ) the radiative decay rate of the spontaneous emission of a chiral molecule located in close vicinity to the surface of a chiral spherical particle ( @xmath205 ) with the size @xmath169 , as a function of the chirality parameter @xmath173
. a dielectric sphere has @xmath206 and @xmath199 , a metal sphere has @xmath207 and @xmath199 , an lh sphere has @xmath207 and @xmath208 .
( a ) the transition electric dipole moment of the molecule is oriented along the @xmath200-axis ( @xmath209 ) , i.e. it is tangential to the surface of the particle .
( b ) the transition electric dipole moment of the molecule is oriented along the @xmath202-axis ( @xmath210 ) , i.e. it is normal to the surface of the particle .
the solid line corresponds to the transition magnetic dipole moment of the molecule oriented along the x - axis , and the dashed line corresponds to the transition magnetic dipole moment of the molecule oriented along the z - axis .
the value of the transition magnetic dipole moment @xmath211 .
the particle is placed in vacuum.,width=340 ] ( color online ) the radiative decay rate of the spontaneous emission of a chiral molecule located in close vicinity to the surface of a chiral lh spherical nanoparticle ( @xmath205 ) with @xmath212 , @xmath213 , @xmath214 , and @xmath215 . ( a )
the electric and magnetic dipole moments of the transition of the molecule are oriented along the @xmath200-axis ( tangentially to the particle surface ) .
( b ) the electric and magnetic dipole moments of the transition of the molecule are oriented along the @xmath202-axis ( normally to the particle surface ) .
the dashed line shows the asymptotic expression ( [ eq46 ] ) .
the nanoparticle is placed in vacuum.,width=340 ] ( color online ) ( a ) the ratio of the effective radiative decay rate of spontaneous emission of `` left '' molecules with @xmath216 to the effective radiative decay rate of spontaneous emission of `` right '' molecules with @xmath217 and ( b ) vice versa as a function of the dielectric permittivity ( @xmath6 ) and the magnetic permeability ( @xmath7 ) of the material from which chiral spherical nanoparticle is made .
the chirality parameter @xmath218 .
the molecule is placed in close vicinity to the surface of the spherical nanoparticle ( @xmath219 ) .
the white line shows the position of chiral - plasmon resonance in the nanoparticle ( see eq .
( [ eq45 ] ) ) .
the nanoparticle is placed in vacuum.,width=340 ] | in the framework of the perturbation theory of the nonrelativistic quantum electrodynamics , a theory of spontaneous emission of a chiral molecule located near a chiral ( bi - isotropic ) spherical particle is developed . it is shown that the structure of photons in the presence of chiral spherical particles differs significantly from the structure of te or tm photons .
exact analytical expressions for the spontaneous emission radiative decay rate of a chiral molecule with arbitrary electric and magnetic dipole moments of transition located near a chiral spherical particle with arbitrary parameters are obtained and analyzed in details .
simple asymptotes for the case of a nanoparticle are obtained . substantial influence of even small chirality on a dielectric or `` left - handed '' sphere is found .
it is shown that by using chiral spherical particles it is possible to control effectively the radiation of enantiomers of optically active molecules . |
the increasing demand for accurate control of quantum devices using high - fidelity control protocols@xcite , has stimulated interest in the study of the dynamics of quantum systems in response to slowly varying hamiltonian .
moreover , rapid progress in the field of adiabatic quantum computing has fueled further interest in and need for more careful analysis of the dynamics of quantum systems whose parameters vary slowly in time.@xcite in addition , decoherence in any real quantum system sets a rigid constraint on the time interval during which a quantum protocol must be carried out , limiting all protocols to intermediate time intervals that are shorter than the decoherence time . at these intermediate time scales , both non - adiabatic corrections and coupling to the environment
become equally important .
the previous analysis@xcite of the qubit dynamics with time - dependent hamiltonians was based on the lindblad master equation@xcite that describes the interaction with environment in terms of dephasing and transition processes characterized by phenomenological decoherence rates . an alternative microscopic approach , formulated as a perturbative theory for a quantum system with a time - independent hamiltonian interacting with its environment , introduces the bloch redfield ( br ) master equation@xcite .
if the hamiltonian of the system changes in time , the br approach has to be modified to properly account for a non - adiabatic corrections . in this paper
we extend the br approach to account for slow evolution of the system hamiltonian in the presence of the environment .
the main concept of the br theory is based on the identification of decoherence processes in terms of the matrix elements for transitions caused by environment in the eigenstate basis of the quantum system.@xcite for the hamiltonian that varies with time , one can still use a basis defined by eigenvectors of the hamiltonian@xcite , where the hamiltonian is always represented by a diagonal matrix @xmath0 , where the unitary transformation @xmath1 denotes a transition from the original basis to the eigenstate basis .
time - dependence of @xmath1 produces an extra term in the time evolution of the quantum system that is effectively described by the new hamiltonian @xmath2 .
this expression is not necessarily diagonal and another basis transformation is required .
such series of diagonalization transformations can be continued indefinitely , but for slowly changing hamiltonian , the series can be truncated after a finite number of transformations neglecting terms of the higher order in time - derivatives of the parameters in the hamiltonian .
in addition to changes in the effective spectrum of the system , matrix elements representing coupling between the quantum system and its environment are also modified , resulting in a redefinition of the transition rates for the system .
we focus our analysis on the dynamics of a two - level quantum system a qubit or a spin-1/2 system in the presence of time - dependent field , which we refer to below as the control field of the qubit .
we study the dynamical response of the transverse magnetization to quench velocity of the control field .
the transverse magnetization measurements can provide the value of the berry curvature of a quantum system @xcite and , consequently , characterize topological properties of a ground state of the system . since any real qubit
is always coupled to its environment , it is necessary to perform detailed analysis of the non - adiabatic dynamics of a qubit system in the presence of dissipation . to this end , we investigate the effect of pure dephasing and energy relaxation due to the ohmic bath on the qubit polarization .
our results indicate that the decoherence suppresses the transient wiggles of the out - of - plane qubit projection , thereby bringing up the linear relation between the qubit response and the quench velocity .
thus , the dissipation facilitates the berry curvature measurement based on the non - adiabatic response , proposed in ref .
@xcite . furthermore
, this study is also related to the measurement technique of the berry phase in qubits , which are based on an interference experiment where the drive parameter was changed slowly @xcite , see also refs .
@xcite for theoretical analysis of the influence of environment on the berry phase .
we also apply the modified br equation to the landau
zener ( lz ) problem@xcite in a qubit coupled to environment at arbitrary temperature .
the lz problem in a quantum system coupled to its environment has attracted significant interest recently , where the environment was considered either as a source of classical noise@xcite , or quantum fluctuations that cause transitions between qubit states @xcite , or pure dephasing@xcite .
more recently , the lz interferometry has attracted a growing interest@xcite .
here we focus on the role of quantum fluctuations in the environment that cause transitions between the eigenstates of the qubit in the lz problem .
we argue that during the lz transition , the matrix elements of the coupling between the qubit and its environment must be considered in the basis of eigenstates of the full qubit hamiltonian and therefore , the matrix elements acquire an explicit time dependence due to rotation of the eigenstate basis in addition to straightforward dependence on the energy difference between the eigenstates .
this treatment modifies the previous results of refs .
@xcite and generalizes the results of refs .
@xcite , where a similar basis transformation was naturally included in the calculations .
we disregard the effect of the lamb stark shift on the qubit spectrum due to coupling to the environment , considered in ref .
@xcite , since this can be included in the redefined control field of the qubit .
we focus solely on the transition effects due to non - unitary evolution of the qubit density matrix .
we consider the quantum fluctuations of the environment that are fixed along the direction of the control field at very long initial and final moments of the lz transition so that the matrix element that characterizes the transition between qubit states at long times is absent and environment produces dephasing only
. for arbitrary direction of the fluctuating field , the transition remains effective over long time and will effectively bring the qubit to the ground state for zero temperature environment .
we also consider `` dephasing '' coupling@xcite when the quantum fluctuations occur only in the direction parallel to the direction of the control field in the parameter space of the qubit hamiltonian .
our result is in agreement with ref .
@xcite of the same problem within lindblad master equation , in the limit of a high - temperature environment .
this paper is organized as follows . in section [ sec2 ]
, we present a formalism of the br equations in transformed basis for time - dependent hamiltonians . in section [ sec3 ] , we study the evolution of a qubit whose control field rotates in a plane with a constant magnitude and consider different directions of the environmental coupling field . in section [ sec4 ] , we consider the lz problem in the presence of zero and finite temperature environment and show that transition is dominated by thermal excitation of the qubit at finite temperatures . in section [ sec5 ] , we analyze the non - adiabatic effects within the lindblad formalism .
we end with conclusions in section [ sec6 ] .
we consider a spin coupled to a bath of harmonic oscillators .
the full hamiltonian @xmath3 is a sum of the hamiltonian for the spin in the magnetic field @xmath4 @xmath5 the interaction hamiltonian of the spin with the environment @xcite @xmath6 and the bath hamiltonian @xmath7 here we assume that each environment oscillator interacts with the spin as a quantized magnetic field @xmath8 in the common direction @xmath9 , @xmath10 and @xmath11 are raising and lowering operators of the field .
the reduced density matrix @xmath12 of the spin is determined by tracing out environment degrees of freedom of the full density matrix @xmath13 .
the full density matrix satisfies the unitary master equation @xmath14\,.\end{aligned}\ ] ] there are several approaches to obtain the corresponding equations for time evolution of the reduced density matrix for the qubit .
here we consider the limit of weak coupling of a qubit to the environment , when the density matrix is defined by the br equations,@xcite see also refs .
@xcite where a diagrammatic technique was developed to treat the weak coupling to environment .
the environmental effects are characterized by the spectral density function of the coupling @xmath15 .
a generic spectral function has a power law dependence on energy at small energies , @xmath16 , and vanishes rapidly for energies above the ultraviolet cutoff @xmath17 . here , we consider the ohmic ( @xmath18 ) environment with exponential high - energy cutoff : @xmath19 where the dimensionless parameter @xmath20 defines the strength of coupling between the qubit and its environment and @xmath17 is the cutoff .
we restrict ourself to the weak coupling limit , @xmath21 . in general , the effect of weak environment on the qubit dynamics is twofold . on one hand
, the qubit hamiltonian is renormalized by the environment modes with @xmath22 , known as the lamb and stark effects . on the other hand ,
when we integrate out the environmental degrees of freedom , we also obtain non - unitary terms in the evolution of the quantum system .
both of these effects are accounted for by the br equation@xcite for the qubit density matrix @xmath23 .
we first consider the case of a constant external magnetic field along @xmath24 direction , @xmath25 .
then , the br equation has the following form in the eigenstate basis [ eq : brall ] @xmath26 we obtained the above equations within secular approximation that neglects fast oscillating terms with frequencies larger than the decoherence rates . the equation in the matrix form , eq . , determines the evolution of diagonal elements of the density matrix .
the relaxation and excitation rates , @xmath27 and @xmath28 , are defined by the spectral density @xmath29 at the energy corresponding to the energy difference between two states of the qubit : [ eq : gamma_re ] @xmath30 and @xmath31 $ ] is the planck s function .
the factor @xmath32 indicates that only the component of the fluctuating environment field that is perpendicular to the direction of the control field @xmath33 gives rise to the qubit flip processes .
the off - diagonal elements of the density matrix are characterized by the decoherence rate @xmath34 and pure dephasing rate @xmath35 given by @xmath36 the decoherence stems from two processes the qubit flip processes with rate @xmath37 , and pure dephasing which is not responsible for energy transitions at low frequency with rate @xmath38 .
the only source of pure dephasing is the fluctuating fields of the environment along the external field @xmath33 , hence the factor @xmath39 in the definition of the pure dephasing term , @xmath40 .
the renormalization of the qubit hamiltonian by the environment due to the lamb or stark effects are determined by the imaginary part of the environmental correlation function , as discussed in ref .
explicitly , the renormalized qubit energy @xmath41 is @xmath42 where @xmath43 denotes the cauchy principal value .
below , we assume that the control field @xmath33 already includes renormalization effects from the environment . the goal of this paper is to investigate the features of the qubit evolution originating from decoherence characterized by rates @xmath44 and @xmath45 , respectively .
the significance of the effect of the lamb and stark shifts on the evolution of the qubit was demonstrated in ref .
@xcite in the context of the lz problem .
we note that the qubit density matrix can be defined in terms of the magnetization in @xmath46 , @xmath47 and @xmath48 directions as @xmath49 then the br equations , eq .
, acquire a more common form of the bloch equations [ eq : blochall ] @xmath50 the above br equations were obtained in the basis of qubit eigenstates . in case
when the control field @xmath4 changes in time , we perform transformation @xmath51 of the basis that keeps the qubit hamiltonian diagonal .
this basis is commonly referred to as adiabatic .
the corresponding transformation has two consequences .
the first consequence of @xmath51 transformation is that the hamiltonian in the new basis acquires an extra term originating from the time dependence of the transformation @xmath51 .
thus , the qubit hamiltonian in the new basis is @xmath52 the resulting hamiltonian still may remain non - diagonal due to the berry connection term , @xmath53 .
we can introduce a new transformation @xmath54 that diagonalizes the right hand side of , but this transformation generates a new term @xmath55 and the `` diagonalization '' series of transformations @xmath56 does not stop for an arbitrary time evolution of @xmath4 , because the berry connection terms appearing in each consecutive diagonalization transformation acquires an extra time derivative .
however , for slow time evolution , the series of transformations can be truncated by the first one or two transformations . since the br treatment of environmental effect requires anyway that the system changes in time slower than the rates given by eqs . and in the master equation , the truncation to a limited number of transformations @xmath56 under slow evolution of @xmath4 is justified .
also , in a special case of constant rotation of @xmath4 in a plane , the second transformation @xmath57 is time - independent and transformation series stops after this second basis rotation .
the second consequence of the basis transformations is the modified interaction term in that the coupling between the qubit and its environment @xmath58 is modified from the initial coupling operator @xmath59 to the environment field by the transformation matrix @xmath60 .
this transformation changes the corresponding `` projection '' factors @xmath61 in eqs . as well as the spectral weights @xmath29 .
modification of the coupling between the qubit and its environment , introduced by eq . ,
swaps components of the fluctuating field responsible for the pure dephasing and transition processes .
for example , in case of a fixed external field @xmath62 , fluctuations along @xmath63 give rise to pure dephasing and do not cause transition processes between qubit eigenstates .
however , as @xmath4 rotates while @xmath9 remains in @xmath63 direction , the fluctuating component along field @xmath4 is the only one responsible for the dephasing with the corresponding rate proportional to the spectral weight of its low - frequency fluctuations @xmath64 , while the component of the fluctuating field perpendicular to @xmath33 will produce qubit flip processes with the rate characterized by the spectral weight of fluctuating field with the energy equal to the energy of qubit flip @xmath65 .
the second unitary transformation further mixes matrix elements of the coupling to environment representing qubit flip processes and pure dephasing .
below , we present explicit expressions for the rates in eqs . for two special cases of evolution of @xmath4 for different types of environment .
we focus on the effect of qubit flip processes due to environment and assume that @xmath66 in most numerical solutions .
we note that the pure dephasing produced by the low frequency noise of the environment can be successfully described in terms of fluctuations of the classical field and may also include non
markovian time correlations that are omitted in the br approach .
effects of classical noise were discussed in refs .
@xcite for the lz transition and in refs .
@xcite for berry phase measurements .
we first consider a qubit with the hamiltonian characterized by a time - dependent field in @xmath67 plane : @xmath68 . by definition ,
@xmath69 for @xmath70 .
the transformation to adiabatic basis is defined by : @xmath71 and the resulting qubit hamiltonian has the form @xmath72 here , the second term is responsible for the non - diagonal form of the hamiltonian for time - dependent rotation angle @xmath73 and causes the resultant field to point out of the rotation plane of @xmath4 .
this hamiltonian has eigenvalues @xmath74 and eigenvectors , which are different from the vectors of the adiabatic basis .
the latter two represent spin states in the ( @xmath67 ) plane with @xmath75 . on the contrary
, the qubit in the ground state @xmath76 of the hamiltonian has a non - zero expectation value of the polarization @xmath77 in the direction perpendicular to the ( @xmath67 ) plane of the control field @xmath33 : @xmath78 in the limit of slow rotations , @xmath79 , this result is consistent with a more general expression that connects a generalized force @xmath80 to time - dependent parameters @xmath81 of the hamiltonian through the berry curvature @xmath82 as @xcite @xmath83 comparing eq . and
eq . , we identify @xmath84 , @xmath85 and @xmath86 .
explicitly , the coefficient of the linear term in the rate of change of the magnetic field , _ i.e. _ @xmath87 , is the berry curvature @xmath88 .
indeed , this value of the berry curvature gives the berry phase @xmath89 for one full rotation of the control field in the @xmath90 plane after its integration over the half - sphere , @xmath91 .
this relation holds for an isolated qubit controlled by field @xmath4 , assuming that @xmath4 is a slowly varying function of time with continuous higher derivatives
. however , if the rotation of the control field @xmath33 starts instantaneously with constant angular velocity @xmath92 , _ i.e. _ @xmath93 , the rotation is equivalent to a quantum quench in the representation of eq . from @xmath94 to @xmath95 .
the qubit that was initially in the ground state of the original time - independent hamiltonian , @xmath96 , is in the superposition of eigenstates of the new hamiltonian and exhibits precession around new direction of the effective field @xmath97 .
this precession causes oscillations of @xmath98 around its average value given by eq . .
in this section we demonstrate that a qubit coupled to a zero - temperature environment relaxes towards the lower eigenstate of hamiltonian and for long time limit after the rotation started , the qubit state obeys eq . . for rotation with constant angular velocity @xmath99 , the transformed hamiltonian , eq
is time independent and can be diagonalized by the second basis transformation @xmath100 the qubit hamiltonian in a new basis after a full transformation @xmath101 becomes fully diagonal with time - independent eigenvalues : @xmath102 we can apply the br equation for the qubit density matrix , where the rates in eq . are defined by the interaction term @xmath103 , eq . , with @xmath104 replaced by its transformation under @xmath105 according to eq . .
the result of the @xmath105 transformation depends on the original orientation of the vector @xmath9 in the qubit space .
below , we consider three orientations of @xmath9 .
we note that for the limit @xmath106 considered in this section , the shift of eigenvalues of hamiltonian and modification of the coupling to environment by the second transformation @xmath107 is not significant and can be disregarded to the lowest order in @xmath99 .
( color online ) time dependence of the out - of - plane polarization , @xmath108 , at zero temperature of environment for @xmath109 ( solid line ) and @xmath110 for fluctuating environment field out of the plane of rotation , @xmath111 .
the pure dephasing rate is zero , @xmath66 .
the frequency of rotation of the control field is @xmath112 . ]
we first consider the case when the coupling between the qubit and its environment is determined by the vector @xmath113 perpendicular to the plane of rotation of the external field @xmath4 . for time
independent hamiltonian , this coupling causes qubit flip processes and the corresponding decoherence rates are defined by the environment spectral function at the excitation energies equal to the qubit energy splitting . for time - dependent
hamiltonian with rotating @xmath4 , we have to write the qubit coupling operator @xmath114 in the rotated basis that diagonalizes the original hamiltonian .
as we discussed above , the transformation is a product of two consecutive transformations .
the first transformation , @xmath51 to the adiabatic basis does not change the coupling operator @xmath115 .
the second transformation results in @xmath116 here , the first term represents the qubit flip process , while the second term preserves the qubit orientation and causes pure dephasing .
the corresponding rates in the br equations are given by [ eq : ratesa ] @xmath117 , \\ \gamma_{e } & = \frac{\cos^{2}\eta}{2}j(w ) n(w),\\ \gamma_{2 } & = \frac{\gamma_{r}+\gamma_{e}}{2 } + \frac{\sin^{2}\eta}{2}j_0,\end{aligned}\ ] ] with @xmath118 and @xmath119 defined by eqs . and .
the qubit dynamics is characterized by the relaxation and excitation rates proportional to the spectral function @xmath120 of environment at energy @xmath118 , these rates appear with factor @xmath121 and recover the case of the qubit with a time - independent hamiltonian with @xmath122 when only environment modes in resonance with the qubit contribute to the qubit dynamics . at finite @xmath99 , however , the pure dephasing mechanism arises after transformation @xmath123 and originates from the low frequency modes of the environment with spectral density @xmath64 .
the pure dephasing rate contains factor @xmath124 which is small for slow rotation with @xmath106 .
( color online ) time dependence of the out - of - plane polarization , @xmath108 , at various temperatures of environment : @xmath125 ( solid line ) , @xmath126 ( dashed line ) and @xmath127 ( dotted line ) for fluctuating environment field out of the plane of rotation , @xmath111 . the pure dephasing rate is zero , @xmath66 .
the frequency of rotation of the control field is @xmath112 .
the coupling to environment @xmath110 . ] the bloch equations eq . with rates given by eq . can be solved to get the qubit density matrix @xmath128 in the secondly rotated basis . in conjunction with the initial condition , the time evolution of @xmath108
is then obtained by @xmath129 $ ] .
first , we provide an exact analytical solution by choosing the initial state to be a thermal state @xmath130 . defining @xmath131
, the initial condition for the bloch equation becomes @xmath132 , @xmath133 and @xmath134 .
integrating the bloch equation with the above initial condition yields @xmath135 where @xmath136 and we assumed @xmath66 . in the long times limit , @xmath137 , @xmath108 reaches its stationary state solution @xmath138 regardless of the form of the initial state .
the significance of this expression is that the dynamical transverse response of the qubit subject to a rotating magnetic field is a consequence of the geometric phase effect in the sense that the stationary value @xmath139 does not depend on the strength of the coupling to environment .
therefore , @xmath139 is purely geometrical and immune to quantum zero - temperature fluctuations of the environment .
next , in order to get the numerical solution of the br equations we utilize standard integration methods for a system of linear differential equations with time - dependent coefficients .
alternatively , we obtain the same results using the br functions of the qutip package @xcite with a proper adjustment to the system hamiltonian and the interaction term , see eqs . and , for time
dependence of the eigenstate basis , as presented in figs .
[ fig:3a_my_vs_t_variosgammatzero ] and [ fig:3a_my_vs_t_for_temp ] .
we verified that the results shown in the plots are identical to numerical integration of the br equations with the rates given by eqs . .
in both plots , the initial condition of the density matrix is chosen to be the ground state at @xmath140 when @xmath141 .
we obtain plots consistent with the analytical result , eq .
, for the thermal state of the density matrix at @xmath140 . in fig .
[ fig:3a_my_vs_t_variosgammatzero ] , we present the time evolution of @xmath108 for several values of the coupling to the environment . from the plot it is clear that the role of the environment is to suppress transient wiggles of @xmath77 and to bring the system to the steady state , defined by eq . with @xmath142 .
however , the transverse magnetization is fragile to thermal fluctuations , since these fluctuations create excitation to the higher energy state .
the result is shown in fig .
[ fig:3a_my_vs_t_for_temp ] , where we fix @xmath20 and plot @xmath108 for different temperatures @xmath143 .
we note that since the dephasing rate , @xmath144 grows with the temperature , the oscillations decay faster for higher temperatures . also , at finite temperatures , the spin has nonzero probability to stay in the excited state , the asymptote of @xmath145 is reduced in agreement with eq . .
( color online ) time dependence of the out - of - plane polarization , @xmath108 , at zero temperature of environment , for @xmath110 ( solid line ) and @xmath146 ( dashed line ) for fluctuating environment field in the plane of rotation , @xmath147 .
the pure dephasing rate is zero , @xmath66 .
the frequency of rotation of the control field is @xmath112 .
the relaxation is reduced for time intervals when @xmath148 . ]
we now consider the qubit interacting with environment field in the plane of rotation .
we take @xmath149 and for @xmath150 the coupling to the environment results in pure dephasing and is characterized by the low frequency spectral density @xmath64 . as @xmath33 rotates , the effect of environment alternates between pure dephasing and qubit transitions between eigenstates .
we obtain this variation in qubit flip and dephasing rates already after applying transformation @xmath151 to the interaction hamiltonian of the qubit and environment , eq . .
however , for rotating @xmath4 we have to take into account the gauge term @xmath152 in eq . by applying the second transformation @xmath123 to @xmath153 .
we obtain @xmath154 that contains matrix elements for qubit flip processes at any moment of time .
the corresponding rates in the bloch redfield equations are [ eq : ratesz ] @xmath155 , \\ \gamma_{e } & = \frac{g(t)}{2}j(w ) n(w),\\ \gamma_{2 } & = \frac{\gamma_{r}+\gamma_{e}}{2 } + j_0 \cos^{2}\eta\cos^{2}\omega t , \end{aligned}\ ] ] where @xmath156 and thus the qubit flip rates are nonzero as a function of time .
the evolution of the qubit in this case corresponds to precession of a spin in the magnetic field with initial state distinct from its new ground state after the quench .
namely , its dynamics will correspond to suppression of off - diagonal elements of its density matrix with the rate @xmath157 and equilibration of the diagonal elements of @xmath158 with rates @xmath159 .
we emphasize that in this case all decoherence rates are time - dependent .
we calculate time - dependence of @xmath108 by numerically solving the br equations with the rates given by eq . .
we present the result of integration in fig .
[ fig:3b_z_zerotemp_variousgamma ] for two different values of @xmath20 at zero temperature and find clear evidence that the decoherence rates are roughly one half smaller compared to the result of previous subsection for the same value of @xmath20 .
meanwhile , in fig . [ fig:3b_z_variostemp ] we fix @xmath20 and plot @xmath108 for different temperatures . at time
longer than the relaxation time @xmath160 , @xmath108 becomes constant with its value @xmath161 , see eq .
( color online ) time dependence of the out - of - plane polarization , @xmath108 , for @xmath146 and @xmath125 ( solid line ) , @xmath126 ( dashed line ) and @xmath127 ( dotted line ) in case when fluctuating environment field is in the plane of rotation , @xmath147 .
the pure dephasing rate is zero , @xmath66 .
the frequency of rotation of the control field is @xmath112 . ]
we also consider a somewhat artificial scenario when the coupling vector @xmath162 in eq
. rotates together with the external field @xmath4 .
e.g. _ when @xmath4 is realized as two quadratures of microwave pulse driving a qubit and the environment is described by longitudinal quantum fluctuations of the pulse . ] for a stationary hamiltonian this environment does not produce qubit flip processes and results in pure dephasing , when the diagonal elements of the density matrix do not change and only off diagonal elements decrease with time . in case when the direction of the control field rotates with frequency @xmath99 , the basis transformation term in eq .
introduces qubit flip processes for this coupling with the rates in eqs . given by [ eq : gammas_lf ] @xmath163 , \\
\gamma_{e } & = \frac{\sin^2\eta}{2}j(w ) n(w),\\ \gamma_{2 } & = \frac{\gamma_{r}+\gamma_{e}}{2 } + \cos^{2}\eta j_0 . \end{aligned}\ ] ] for slow rotation @xmath106 , we have @xmath164 and qubit flip processes are small . in this case , dephasing will suppress precession on time scale @xmath165 , and further equilibration of the system occurs on a longer time scale @xmath166 .
we describe the evolution of a qubit coupled to high temperature environment using a dephasing lindblad model in sec .
[ sec5 ] .
( color online ) time dependence of the out - of - plane polarization , @xmath108 , at zero temperature of environment for a qubit coupled to a damped harmonic oscillator with damping rate @xmath167 and coupling constant between the qubit and environment @xmath168 . coupling vector @xmath169 ( solid line ) and @xmath170 ( dash - dotted line ) . for comparison ,
the solution for the bloch redfield equation is presented ( dashed line ) with @xmath171 and @xmath66 .
the rotation angular velocity is @xmath112 . ] in this subsection we consider the interaction of a qubit with a single damped quantum harmonic oscillator .
this model can be used to describe environment with a sharp spectral function @xmath29 .
the interaction part of the hamiltonian is similar to eq . :
@xmath172 and the single - mode hamiltonian of the oscillator is @xmath173 .
we describe dissipation of the oscillator using the lindblad relaxation operators for the full density matrix @xmath174 of the qubit and the oscillator system : @xmath175 -\kappa \left ( \hat{a}^\dagger \hat{a } \bar \rho+ \bar \rho \hat{a}^\dagger \hat{a}- 2 \hat{a}\bar \rho \hat{a}^\dagger \right)\ ] ] this equation is a standard lindblad master equation with time dependent hamiltonian .
the difference with the previous calculations of this section is that we keep a full quantum mechanical treatment of the qubit interaction with the oscillator and perform all transformations of the qubit basis for the full hamiltonian of the qubit and the oscillator . at the same time , we assume that the lindblad superoperator for the relaxation of the harmonic oscillator , represented by the last term in eq .
, is not affected by these transformations .
we evaluate the qubit projection perpendicular to the rotation plane of the control field as a function of time .
[ fig:3c_zerotemp ] shows the comparison between calculation of bloch - redfield equations and damped quantum oscillator with different coupling directions at zero temperature .
all three curves saturate at universal value @xmath176 .
it is worth pointing out that the @xmath170 coupling results in time - dependent transition rates that are at minimum when @xmath177 and at maximum when @xmath122 , as one can conclude from the amplitude of oscillations of @xmath108 for @xmath178 .
effectively , the overall relaxation is slower than that of the case @xmath169 and the amplitude of oscillating @xmath77 at @xmath179 decays insignificantly .
the calculations at finite temperature @xmath180 are plotted in fig .
[ fig:3c_temp0p5 ] and in all cases @xmath139 is consistent with eq . .
( color online ) time dependence of the out - of - plane polarization , @xmath108 , at environment temperature @xmath126 for a qubit coupled to a damped harmonic oscillator with damping rate @xmath167 and coupling constant between the qubit and environment @xmath168 . coupling vector @xmath169 ( solid line ) and @xmath170 ( dash - dotted line ) . for comparison ,
the solution for the bloch redfield equation is presented ( dashed line ) with @xmath171 and @xmath66 .
the rotation angular velocity is @xmath112 . ]
in this section we consider the landau zener transition in a qubit coupled to its environment .
the external field in the qubit hamiltonian has the following form @xmath181 , where @xmath182 is the minimal level separation and @xmath183 characterizes the rate at which the hamiltonian changes . for the landau
zener problem , the qubit is initially in the ground state @xmath76 with the density matrix @xmath184 . the task is to find the probability of the system to be in the excited state @xmath185 which is given by @xmath186 .
effects of the environment on qubit s dynamics can be separated into pure dephasing of the qubit state during the lz process and inelastic qubit flips .
when we consider a qubit coupled to its environment that causes qubit flip processes , we have to be careful with the formulation of the lz problem .
indeed , the lz process is formally infinitely long and the qubit flip processes accompanied by the energy exchange will result in equilibration of the qubit system with its environment . in particular , for the zero temperature environment , the qubit will relax to the ground state even if it was temporarily excited during the lz process .
for environment at finite temperature , the qubit state will tend to thermal state @xmath187 with @xmath188 .
but as formally @xmath189 for long times @xmath190 , the qubit will relax to the ground state and we find @xmath191 .
previous considerations , see _ e.g. _
@xcite , predicted @xmath192 for the ohmic environment with large high - frequency cutoff in the environment modes .
but in this case the problem looses its meaning since the lz transition is shadowed by trivial relaxation of a quantum system to its ground state by releasing its energy to the environment .
one can reformulate the problem in terms of finite time lz process , which may be experimentally relevant situation in some cases .
alternatively , one can assume that the environment spectral function has a relatively low cutoff at high frequencies @xmath193 and the relaxation is absent after time @xmath194 . here
, we consider a special orientation of the coupling vector with environment when @xmath195,where @xmath63 is defined by @xmath196 .
in this situation , the relaxation processes becomes weak at long times @xmath197 .
this type of coupling is expected to be dominant in qubits with relatively long energy relaxation times , but with short dephasing time due to dominant coupling with the fluctuating field parallel to the qubit field along @xmath63 .
we utilize the bloch
redfield approach to the problem of landau zener transitions in the presence of environment with @xmath198 . in principle , we need to write the br equations in the basis where the transformed qubit hamiltonian is diagonal after an infinite series of basis transformations given by @xmath199 , which can be an infinite series .
however , under the condition @xmath200 , the series of basis transformations can be limited by @xmath201 .
( color online ) projection of a qubit state during the landau
zener process on the bloch sphere on @xmath202 state in the diabatic basis ( dash - dotted line ) and on the ground state in the adiabatic basis ( solid line ) and the `` improved '' eigenstate basis ( dashed line ) . in the diabatic basis
the projection of the qubit state shows long oscillations with amplitude decreasing as a power law in time , while the eigenstate projections quickly reach the asymptotic value .
level - crossing speed @xmath203 and no coupling to environment . ]
the first transformation changes the representation from diabatic basis of states @xmath204 and @xmath202 along @xmath63 to the adiabatic basis of the ground , @xmath76 , and excited , @xmath185 , states , where the hamiltonian is diagonal .
the first transformation matrix @xmath51 has the same form as in eq . except the rotation angle @xmath73 , which is now defined as @xmath205 the transformed hamiltonian in the adiabatic basis has the form@xcite @xmath206 the second transformation is chosen to diagonalize matrix @xmath207 and has the form @xmath208 the hamiltonian in this `` improved eigenstate '' basis has the form [ eq : h0u2lz ] @xmath209 without dissipation , the lz problem is equivalent in all three representations , with a properly written hamiltonian , i.e. , eq . for the diabatic basis , eq . for the adiabatic basis , and eq
. for `` improved eigenstate '' basis . in all representations
, the qubit follows the appropriate instantaneous control field @xmath210 , but since this field is time - dependent , the qubit deviates from the instantaneous direction of @xmath210 and acquires an additional precession around the control field . when the original field eventually reaches its final direction , @xmath150 at @xmath211 , the direction of the control field becomes time independent and the qubit simply precesses around @xmath63 with a non - zero projection of its state on the excited state , given by the known expression@xcite @xmath212 note that in fig .
[ fig:3spheres ] this precession remains in all three considered representations , but the overall trajectories are smoother in the transformed representations .
as we look at the projection of the qubit state on the `` excited state '' @xmath213 in the appropriate basis , see fig .
[ fig : lzidealpz ] , the oscillations decrease faster in the transformed representations , because the control field @xmath4 aligns faster with its final direction .
we also note that since the control field remains aligned with its initial direction longer in transformed basis , the numerical computation can run over shorter time intervals thus making computation faster and more accurate .
next , we take into account interaction with the environment within the bloch redfield approach .
the coupling to the environment is modified in the diagonal basis of the hamiltonian , see eq . and ref
. @xcite . under the markovian approximation and to the second order in the coupling to environment , we obtain the corresponding br equations in the form [ eq : brlz ] @xmath214 where @xmath215 and @xmath216 are given by eq . .
the rates for the above equations are [ eq : brrateslz ] @xmath217 , \label{eq : gammarlz } \\ \gamma_{e } & = \frac{g_{lz}(t)}{2}j(w(t ) ) n(w(t)),\\ \gamma_{2 } & = \frac{\gamma_{r}+\gamma_{e}}{2 } + j_0 \cos^{2}\eta\cos^{2}\theta(t ) , \end{aligned}\ ] ] where @xmath218 is a function of time
dependent basis rotation angles @xmath73 and @xmath219 defined by eqs . and .
we note that the above equations for br rates are given by truncation of transformation series of interaction hamiltonian , eq . , up to the second order , @xmath220 .
therefore , the rates are defined within @xmath221 accuracy .
the unitary evolution described by either @xmath207 or @xmath222 has no approximations and is valid for arbitrary values of @xmath183 .
we emphasize that once the basis transformation gives rise to non - zero decoherence rates , the qualitative results are similar regardless of our choice of the br rates in the basis obtained after either @xmath223 or @xmath224 transformations .
the rates in the @xmath223 basis are given by eq .
with @xmath225 .
we now discuss solution of eq . .
( color online ) the probability of occupation of the excited state in the landau zener transition in the @xmath226 basis .
the temperature of environment is zero , @xmath125 , the level velocity is @xmath203 .
we assume that the dephasing is absent , @xmath66 . the asymptotic curve for @xmath227 is given by eq . with a proper choice of integration constant @xmath228
] we first consider the zero temperature environment and set @xmath229 to focus solely on qubit flips rather than dephasing .
we numerically integrate the br equation and plot the probability of the system to be in the excited state @xmath230 as a function of time in fig .
[ fig : lz_t0 ] for @xmath110 . for numerical integration , we used both direct integration of linear differential equations and the qutip s package for numerical solution of the bloch redfield equations @xcite , obtaining identical results . as the qubit levels go over the avoided crossing , the probability of the qubit to be in the excited state increases , roughly following the same function of time as @xmath231 for an isolated qubit , @xmath232 . as the levels further depart from each other
, the relaxation of the qubit from the excited state becomes the dominant process in the qubit dynamics , and @xmath231 monotonically decreases and becomes constant once the level separation @xmath233 exceeds the ultraviolet cutoff @xmath17 , or @xmath194 and the qubit is effectively decoupled from the environment . in fig .
[ fig : lz_t0 ] we compare the behavior of @xmath231 for different values of @xmath17 . for finite ultraviolet cutoff @xmath234 , the probability @xmath231 saturates for @xmath235 . for @xmath236 ,
the probability @xmath231 slowly decreases for all @xmath237 . to evaluate this suppression
, we can utilize eqs . in the asymptotic regime for @xmath238 ,
when @xmath239 .
we write [ eq : rho11lzt0 ] @xmath240 where we used the relaxation rate @xmath27 from eq . .
the latter equation demonstrates that even for environment with @xmath178 , the relaxation on long times scales is important .
formally , the power law dependence of @xmath231 on time originates from the slow converging integral @xmath241 due to linearly increasing environment spectral function @xmath29 with energy . with a proper choice of integration constant c ,
we obtain a good agreement between computed @xmath231 in fig .
[ fig : lz_t0 ] and asymptote , defined by eq . .
this power
law dependence stops and reaches a fixed value @xmath242 when the qubit level separation exceeds the environment ultra - violet cutoff at times @xmath194 .
we evaluate the long time asymptotic value of @xmath243 by taking into account the high energy cutoff in the environment spectral function , eq . .
we obtain @xmath244 where @xmath245 is the euler s constant , the integration constant @xmath246 and factor @xmath247 describes suppression of the excited state due to slow relaxation while qubit level separation increases from its minimum @xmath182 to values above the cutoff energy @xmath17 , see appendix a for the derivation of eq .
. equations are valid for @xmath248 . for larger values of @xmath20
, one has to take into account the renormalization of qubit hamiltonian when the off - diagonal matrix element in the original hamiltonian @xmath249 is given by the following self - consistent relation@xcite @xmath250 with solution @xmath251 .
hence the relaxation rate is@xcite @xmath252 where @xmath253 is the gamma function . the integration over time with @xmath254 gives@xcite @xmath255 notice that in the limit @xmath248 , @xmath256 , the relaxation rate @xmath27 reduces to @xmath257 in agreement with the relaxation rate in eq . . similarly , eq
. becomes eq .
( color online ) the probability of occupation of the excited state , @xmath231 in the landau
zener transition in the @xmath226 basis at finite temperature of environment for @xmath203 , @xmath110 and @xmath66 .
the solid lines represent solutions of rate equations that show good agreement with the br equations at higher temperatures . ] at finite temperatures , the excitation and relaxation rates may exceed @xmath258 terms for strong enough coupling of the qubit to its environment and slow drive @xmath183 . in this case , we disregard @xmath216 terms in eq . and
the diagonal elements of the density matrix satisfy the rate equations .
since the rate equations preserve the trace of the density matrix , @xmath259 , with @xmath260 , we introduce @xmath261 and obtain the differential equation for @xmath262 : @xmath263 the initial condition is @xmath264 for @xmath265 . while we can write a formal solution to eq .
, the solution is not well defined due to logarithmic divergence of @xmath266 for the spectral function @xmath267 without a cutoff .
we present the result of numerical solution of eqs . and the rate equations in fig .
[ fig : lz_temprateeq ] .
we notice that for higher temperatures , these two solutions are indistinguishable because the thermal effects dominate only in short time scales @xmath268 such that the time window is long enough for the qubit to be thermalized and its off - diagonal elements of density matrix vanish . integrating eq . over @xmath190 yields the following solution of @xmath269
: @xmath270 the integral over time @xmath190 is understood as thermal activation processes with rate @xmath271 and integral in the exponent can be considered as contribution of relaxation processes after thermalization . for weak coupling @xmath248 and not very high temperatures @xmath272
, the integral in the exponential is a slow function of @xmath190 .
therefore , we can replace the lower bound of the integration by @xmath140 .
we obtain @xmath273 in the limit of low temperatures @xmath274 @xmath275 and in the limit of higher temperatures @xmath276 @xmath277 where @xmath278 is defined by eq . .
the details of the derivation of the above equations are presented in appendix [ app : a ] .
we remind that eqs .
are valid when the rate equations are a good approximation to the br equations . in this case , the transition of the system to the excited state is a consequence of incoherent excitation by environment of the qubit , and is not the coherent phenomenon that leads to the excitation in the landau
zener transition of an isolated quantum system .
however , the excitation processes only happen when the adiabatic eigenstates of the qubit have a non - zero matrix elements with the coupling to environment , the latter happens when the `` control field '' @xmath279 is not parallel to the environment field which happens during time @xmath280 , when the excitation rate can be estimated as @xmath281 , resulting in the excitation probability @xmath282 , _ cf .
_ to eq .
as the level separation @xmath283 exceeds temperature , only relaxation process remains that causes transitions to the ground state .
the effect of this relaxation is represented by the exponential factor in eqs . and , _
_ to eq .
( color online ) transition probability @xmath242 as a function of environment temperature @xmath284 , at different values of coupling between the qubit and the environment for @xmath285 .
level - crossing speed @xmath203 , the high energy cutoff for the environment is @xmath286 and @xmath66 .
we take @xmath287 . ] from the above analysis , we conclude that a finite temperature of the environment leads to the `` equilibration '' between the ground and excited states of the qubit , and as temperature increases , the probability of the transition to the excited state in the lz process increases monotonically , cf
this behavior is demonstrated in fig .
[ fig : pinfty_t ] , where @xmath242 is shown as a function of @xmath284 for several values @xmath20 of coupling between the qubit and its environment .
we also note that the temperature effects appear at @xmath288 , at smaller @xmath284 , values of @xmath242 are characterized by the excitation through unitary evolution with the subsequent relaxation .
when we consider @xmath242 as a function of coupling @xmath20 for several values of @xmath284 , we observe a more complicated behavior .
for @xmath125 , shown by the solid line in fig .
[ fig : pinfty_alpha ] , the transition probability @xmath242 monotonically decreases from its value @xmath289 , eq .
, as @xmath20 increases , in agreement with eq . . at finite temperatures , @xmath242 increases for smaller values of @xmath20 , as the excitation process becomes more efficient and provides extra boost for transitions to the excited state in addition to that produced by unitary dynamics
. however , this boost is only a linear function of @xmath20 , see eqs . and , and at stronger values of @xmath20 the exponential dependence of @xmath278 on @xmath20 results in decreasing @xmath242 as @xmath20 increases .
( color online ) transition probability @xmath242 as a function of the coupling parameter of the qubit and the environment , @xmath20 , at different environment temperatures for @xmath285 .
level - crossing speed @xmath203 , the high energy cutoff for the environment is @xmath286 and @xmath66 .
we take @xmath290 . ] as a function of environment temperature @xmath284 for @xmath291 , at different values of drive velocity .
the high energy cutoff for the environment is @xmath286 and @xmath66 .
the solid lines represent solutions of rate equations eq . .
we take @xmath287 . ]
we also consider the environment that produces fluctuating field along the direction of the control field , @xmath292 , in the landau zener problem .
the decoherence rates in the br equations are given by [ eq : gammarlzlong ] @xmath293 , \\
\gamma_{e } & = \frac{\sin^2\eta } { 2}j(w(t ) ) n(w(t)),\\ \gamma_{2 } & = \frac{\gamma_{r}+\gamma_{e}}{2 } + j_0 \cos^{2}\eta.\end{aligned}\ ] ] for this configuration of coupling between the qubit and environment , the matrix elements for transitions between different eigenstates of the qubit caused by the environment are small and the qubit flip rates @xmath294 are proportional to @xmath295 and vanish fast for @xmath296 as @xmath297 . such fast decrease of the qubit flip rates in time simplifies either numerical or analytical integration of the br equation and makes @xmath242 independent from the high - energy cutoff @xmath17 . in particular , for finite temperatures , when the br equations can be reduced to the rate equations , time evolution of @xmath261
is given by eq . with @xmath298 .
the general solution of the rate equation takes similar form to eq . :
@xmath299 performing time integration in eq . gives for @xmath300 : @xmath301 for high temperatures , @xmath302 , we obtain ( see appendix b ) @xmath303.\ ] ] as we mentioned above , the results in eqs . and
are independent from the cutoff energy @xmath17 .
equation shows that @xmath273 vanishes in the low temperature limit , unless we take into account non - adiabatic unitary evolution of the quantum state in the lz problem . in the limit of high temperatures @xmath276 , but still weak coupling , @xmath304 , we obtain the linear dependence of @xmath273 on @xmath284 : @xmath305 which follows from eq . . since simple form of @xmath273
can not be obtained in the intermediate temperature regime , we numerically calculate the solution of rate equation as well as that of bloch - redfield equation for comparison , see fig .
[ fig : pinfty_vs_t_long ] .
when the level - crossing speed @xmath183 is small enough , the transition is mainly due to thermalization at short times and energy relaxation at longer times . in this regime ,
the rate and br equations are in a very good agreement , as demonstrated in fig .
[ fig : pinfty_vs_t_long ] for @xmath306 .
however , as the level crossing speed increases , the non - adiabatic unitary evolution also contributes to the transition to the excited state increasing the probability for a system to be in the excited state .
since the non - adiabatic unitary evolution is not incorporated in the rate equations , the equations underestimate the probability of the excitation in the lz process , compare the solid and dashed curves in fig .
[ fig : pinfty_vs_t_long ] for @xmath203 .
we compare the results obtained from the br equations in the case of longitudinal coupling with the theory based on the lindblad equation for pure dephasing operators . for both problems
, the qubit hamiltonian can be parametrized by the control field @xmath307 , where @xmath283 is the magnitude of the control field equal to the qubit level separation .
the corresponding equation for the density matrix in the adiabatic basis has the form : [ eq : l - lz ] @xmath308+\frac{i\dot \theta}{2}[\sigma_y,\rho]+\frac{\gamma}{2 } ( \sigma_z\rho\sigma_z-\rho).\ ] ] in the component form the above equation is @xmath309 these equations are similar to eqs . , but because they are not written in the eigenstate basis , the last two equations contain extra terms .
time derivatives of diagonal terms contain the off - diagonal terms of the density matrix multiplied by the quantity characterizing the off - diagonal part of the hamiltonian , @xmath310 .
time derivatives of the off - diagonal components of the density matrix have the terms identical to those in eqs . and the extra terms characterized by the diagonal matrix elements and parameter @xmath310 .
in this section we again consider the two cases : ( 1 ) the qubit rotation with a constant angular velocity @xmath311 , i.e. @xmath312 , and @xmath313 ; ( 2 ) the lz problem with @xmath314 and @xmath315 .
when the control field rotates in @xmath90 plane , @xmath316 , the effective hamiltonian is time independent . to make a comparison with the calculation of br equations
, one can look for a quasi - stationary state solution of the density matrix at time scale @xmath317 with ansatz that the off - diagonal elements are @xmath318 .
we disregard @xmath319 terms for @xmath320 and take @xmath321 .
then , we have @xmath322 , @xmath323 and the out of plane qubit projection is@xcite @xmath324 we argue , however , that the above expression does not hold for authentic steady state , @xmath325 , at longer times and for general configuration of the initial conditions .
we present the result of numerical integration of the lindblad equations in fig .
[ fig : my - l ] for @xmath112 and @xmath326 . in our calculation
, we consider the case when the qubit is prepared in the ground state prior to rotation for @xmath70 .
when the rotation starts , the hamiltonian acquires extra terms @xmath327 and the qubit exhibits a precession around new direction of the control field .
this precession is reduced by the decoherence with rate @xmath328 and the oscillatory component in @xmath108 vanishes for times @xmath317 . at longer times
, the diagonal matrix elements start changing as well and the system will eventually relax to @xmath329 and @xmath330 .
the reason for this behavior is that at long times , the diagonal elements acquire significant changes even though these changes have small factor @xmath319 . in the language of the br equation ,
the lindblad pure dephasing operator contains relaxation and excitation components in the eigenstate basis of the transformed hamiltonian @xmath331 and @xmath332 , which is the high temperature limit because it does not distinguish processes with absorption or emission of environment excitations .
correspondingly , the density matrix reaches the high - temperature limit with equal probabilities of occupation of eigenstates of the qubit hamiltonian @xmath333 this asymptotic behavior is consistent with the result obtained from the numerical solution of the lindblad equation , shown in fig .
[ fig : my - l ] .
( color online ) polarization @xmath108 as a function of time @xmath190 for dephasing lindblad evolution . the decoherence rate @xmath326 and rotation velocity @xmath112 .
after the rotation starts , polarization shows an oscillatory behavior originating from the qubit precession , at longer times the precession stops and the qubit relaxes to unpolarized state according to eq . .
] ( color online ) transition probability @xmath242 as a function of daphasing rate @xmath334 for different level crossing speeds @xmath335 .
solid lines are numerical solution of the lindblad equation , eq . , and dashed lines are given by eq . . ]
the expression for landau zener problem to the lowest order in @xmath183 can be obtained from the explicit form of the lindblad equation with @xmath283 given by eq . and @xmath336 .
we assume that the changes in the system are slow and disregard @xmath337 and @xmath338 in eqs .
. then we find @xmath339 $ ] and @xmath340^*$ ] .
substituting these expressions to eq .
, we obtain : @xmath341 \nonumber \\ & = \frac{1}{2}\left[1-\exp\left(-\frac{\pi v}{2\delta^2}r\left(\frac{\gamma}{\delta}\right)\right)\right],\end{aligned}\ ] ] where @xmath342 in the limit @xmath343 , we recover the result of ref .
@xcite : @xmath344 at small decoherence rate and slow drive , @xmath345 , we take @xmath346 and reproduce the previous result , eq . , if we identify @xmath347 .
the agreement between eqs . and has a simple interpretation . the lindblad equation can be viewed as the high temperature limit of the br equation for the ohmic environment@xcite .
the lindblad equation is written in the basis that does not completely diagonalize the hamiltonian operator , and when we rewrite this equation in the basis diagonalizing matrix @xmath348 , we arrive to the collapse operators that represent transition processes between the eigenstates with equal excitation and relaxation rates @xmath349 .
it is the excitation processes that cause transitions of the system to the excited state with the population of an excited state @xmath242 in accordance with eq . .
to account for finite temperatures , the lindbladian operators are to be written in the eigenstate basis of the `` dressed '' hamiltonian , see ref .
. large decoherence rate , @xmath350 , suppresses the off - diagonal elements of the density matrix , and effectively reduces the excitation and relaxation rates @xmath351 . as a result ,
the qubit is more likely to stay in its ground state without experiencing an excitation during the lz avoided level crossing .
the maximum of @xmath352 is reached at @xmath353 .
we compare eq .
( dashed lines ) with the result of numerical integration of the lindblad equation ( solid lines ) in fig .
[ fig : l - lz ] .
we observe that at stronger decoherence rate , when the off - diagonal unitary terms in the evolution of the density matrix can be neglected in comparison with the decoherence terms , @xmath354 in the lz problem , the two solutions are equivalent .
in conclusion , we have presented a detailed analysis of the dynamics of an open quantum system in the presence of time - varying control field .
specifically , we applied the bloch - redfield formalism to a spin-1/2 system whose hamiltonian varies slowly with time and investigated two problems . in the first problem , we studied the response of a qubit to a rotating control field of the qubit with a fixed magnitude .
we noted that when the qubit basis is transformed to keep the effective hamiltonian in the diagonal form , which is required for proper perturbative analysis of the coupling between the qubit and its environment , the transformed hamiltonian acquires extra gauge terms .
the gauge terms result in the modification of the qubit environment coupling and are related to the renormalization of the mass and friction terms due to changing parameters of the hamiltonian , cf .
the exact form of the renormalization depends on a particular orientation of the control field with respect to the fluctuating environment field .
we have illustrated this scenario by considering different orientations of the environment field : ( 1 ) control field and fluctuations are always perpendicular to each other , and the corresponding relaxation rates are time - independent ; ( 2 ) control and fluctuation fields are parallel only at some moments of time , in which case the relaxation rates significantly oscillate in time ; ( 3 ) fluctuations are always along the direction of the control field , then the relaxation rates are small in the parameter given by the ratio of the rotation velocity and level separation .
our analysis offers a clear evidence of robustness of topological features against external noises . to see
this one needs to consider a long time limit where the qubit density matrix reaches a steady state solution that at zero temperature coincides with the ground state of the effective hamiltonian .
when this ground - state qubit configuration is looked at in the original laboratory basis , the qubit has a constant projection in the direction perpendicular to the plane of rotation and the magnitude of the projection is proportional to the product of rotation velocity of the control field and the berry curvature of the qubit ground state . in the long time limit , this response is unaffected by the environmental coupling field , at least for zero temperature environment .
this relation of the response at long times and the berry curvature can be utilized as a practical method for measurements of the chern number@xcite of a quantum system .
we also considered an environment with a very sharp spectral function .
we represent this environment by a quantum harmonic oscillator that has internal relaxation . in this case
we solve the lindblad master equation for the system of coupled qubit and oscillator and find that the results are qualitatively similar to the solution of the br equation with properly chosen relaxation rates . in the second example
, we revisited the landau zener problem .
in this case , the modification of the matrix elements for transitions between eigenstates of the qubit hamiltonian is essential , even though it was not always taken into account.@xcite the eigenstate basis that is necessary to use in treatment of interaction of the qubit with its environment is also convenient for numerical evaluation because in this basis the system behavior during the landau
zener level crossing is represented by a smooth function that quickly reaches its long - time asymptotic value . for a qubit
weakly coupled to the environment , the evolution , long after the level crossing , reduces to suppression of the off - diagonal elements of the density matrix and relaxation of the excited state to the ground state , the latter is accurately described by the rate equations . for the fluctuating field along the asymptotic direction of the control field , the relaxation rate decreases as the level separation increases due to suppression of the matrix elements of qubit transition between eigenstates caused by the environment . however , this suppression is not sufficient to cut the relaxation in the long time limit , and the relaxation results in a power law decay of the excited state , until the separation between the qubit states exceeds the ultra - violet cutoff of the environment . at finite temperature ,
in addition to enhancement of decoherence rates for the qubit , the excitation processes produce transitions from the ground to the excited qubit states , eventually increasing the probability for the qubit to appear in the excited state after the transition .
the br equations accurately describe the crossover for the landau zener transition in an isolated quantum system , eq .
, with unitary evolution , to the open system at arbitrary temperature , see sec .
[ sec4 ] . furthermore , we compare the results obtained from the generalized br equations with that from the lindblad master equation . in particular , we focused on the case of pure dephasing lindblad superoperators,@xcite that are equivalent to the longitudinal coupling of the environment ( fluctuating field of the environment is along the control field ) .
we found that the two results are consistent in the high temperature limit , when the lindblad and br equations are equivalent , but application of the lindblad equation for a system coupled to low temperature environment may result in unphysical solutions .
finally , we note that the generalization of the bloch
redfield equations can be applied to accurately evaluate the fidelity of quantum gates . by taking into account proper modification of the transition and dephasing rates caused by time - varying parameters in the hamiltonian ,
optimization techniques for gate operations can be further improved .
similarly , the br equations for time - dependent hamiltonian are also required for accurate description of protocols for adiabatic quantum computing and the berry phase measurement in recent experiments .
@xcite we thank i. aleiner , a. glaudell , f. nori , a. polkovnikov , s. shevchenko and a. levchenko for fruitful discussions .
the work was supported by nsf grants no .
dmr-1105178 and dmr-0955500 , aro and lps grant no .
w911nf-11 - 1 - 0030 .
here we evaluate the integral in eq . .
notice that while the integral over @xmath355 in the exponent , @xmath356 originates on long interval from @xmath357 to @xmath358 , the second integral converges for time @xmath359 , for not very large temperatures , we can replace the low limit of integration in eq . by @xmath140 . in this case
, we have @xmath360 where @xmath361 , @xmath362 , @xmath363 with @xmath364 and @xmath365 .
first , let us change the integration variable @xmath366 such that @xmath367 and the integral in the exponential then reads @xmath368 this integral can be evaluated in two cases .
first , we consider the low temperature limit @xmath369 , in which the hyperbolic cotangent @xmath370 . therefore ,
the integral is obtained @xmath371,\end{aligned}\ ] ] where @xmath372 is the 0th order modified bessel function of the second kind with the following asymptotes : @xmath373 for @xmath374and @xmath375 for @xmath376 , @xmath377 is the euler constant . as the result , for @xmath274
, we have @xmath378.\ ] ] the first term can be disregarded for @xmath274 . at higher temperatures , there is a stronger contribution to @xmath379 originating from short time interval @xmath359 . we can estimate this contribution as @xmath380 we emphasize that this is the contribution which we do not evaluate correctly when replace eq . by eq . .
therefore , we can treat the above expression for @xmath381 as the boundary of applicability of our approximation , indicating that transition from eq . to is justified not for very high temperatures , such that @xmath382 .
next , we evaluate the integral @xmath383 as before , we first consider the low temperature limit , @xmath274 , in which we approximate @xmath384\simeq \exp(-s / t)$ ]
. then the integral becomes @xmath385 in the high temperature limit , we utilize @xmath384\simeq t / s$ ] , and we obtain @xmath386 this equation is valid for high temperature limit @xmath276 , provided that our substitution of eq . by is justified , or @xmath272 . to sum up
, we evaluated @xmath273 in the limits of low and moderately high temperatures .
the results are presented by eqs . and .
for the longitudinal coupling , the transition probability @xmath242 in limit of low temperatures @xmath387 can be evaluated similarly to the calculations in appendix a. we replace eq . , where the integral over time @xmath190 converges fast for @xmath359 , by the following expression @xmath388 where in the last integral we take the lower limit of integration to zero and @xmath389 . in the above expression , @xmath361 , @xmath362 , @xmath363 with @xmath390 and @xmath365 .
similarly , let us change the integration variable @xmath366 such that @xmath367 .
the integral @xmath391 then reads @xmath392 we note that this integral converges fast and the high - energy cutoff of the environment can be omitted .
similarly , the integral over @xmath393 can be rewritten as @xmath394 in the high temperature limit , we follow a different approach .
we assume that the environment is at high temperature and the relaxation rates are enhanced by factor @xmath395 . in this case , we also have a fast convergence of integrals @xmath396 at @xmath397 and for @xmath276 , we can simplify the rate equation to @xmath398 this equation can be integrated to find @xmath262 with initial condition @xmath399 , and used to define @xmath400 : @xmath401 for @xmath343 , we obtain @xmath402 arriving to eq . . | we study the dynamics of a two - level system described by a slowly varying hamiltonian and weakly coupled to the ohmic environment .
we follow the bloch
redfield perturbative approach to include the effect of the environment on qubit evolution and take into account modification of the spectrum and matrix elements of qubit transitions due to time - dependence of the hamiltonian .
we apply this formalism to two problems .
( 1 ) we consider a qubit , or a spin-1/2 , in a rotating magnetic field .
we show that once the rotation starts , the spin has a component perpendicular to the rotation plane of the field that initially wiggles and eventually settles to the value proportional to the product of angular rotation velocity of the field and the berry curvature .
( 2 ) we re - examine the landau
zener transition for a system coupled to environment at arbitrary temperature .
we show that as temperature increases , the thermal excitation and relaxation become leading processes responsible for transition between states of the system .
we also apply the lindblad master equations to these two problems and compare results with those obtained from the bloch redfield equations . |
the _ @xmath6-core _ of a graph is the largest subgraph with minimum degree at least @xmath6 .
study the following algorithm for finding the @xmath14-core of a graph : if vertices of degree one exist , select one and remove the edge incident to it .
this may cause the degree of other vertices to drop
. if there are no degree one vertices remaining , stop .
repeat .
the graph obtained at the conclusion of this algorithm is the @xmath14-core .
this algorithm is a special case of another , run on hypergraphs , called _ hypergraph collapse _ and first studied in . by a _
we shall mean a map @xmath15 , where @xmath0 is a finite set of _ vertices _ and @xmath16 is the set of subsets of @xmath0 .
it will sometimes be helpful to think in terms of an _ edge - labelling _ of @xmath17 , which is a choice of a set @xmath18 and a map @xmath19 such that @xmath20 for all @xmath21 .
thus @xmath22 describes a set of labelled subsets of @xmath0 , which we call _ hyperedges _ and then @xmath17 gives the number of hyperedges at each subset of @xmath0 .
hyperedges of unit cardinality are called _
patches_. hypergraph collapse is the following algorithm : if a patch exists , select one and remove it together with the unique vertex @xmath23 it contains .
this will cause any other hyperedge @xmath24 containing @xmath23 to be replaced by @xmath25
. if there are no patches remaining , stop . repeat .
although we have described the algorithm in terms of an edge - labelled hypergraph , the possible moves for @xmath17 do not depend on the edge - labelling chosen .
the vertices which are removed by hypergraph collapse are called _ identifiable _ , and hyperedges which contain only identifiable vertices are also called _
identifiable_. these definitions do not depend on the order in which patches are chosen during hypergraph collapse ; see .
the core - finding algorithm of is hypergraph collapse applied to the _ dual _ hypergraph . to obtain the dual , note that we can think of @xmath22 as a subset of @xmath26 .
the roles of @xmath0 and @xmath18 are now symmetric , so @xmath22 also corresponds to an edge - labelling of a hypergraph @xmath27 in which the status of vertices and hyperedges is reversed .
vertices ( resp .
hyperedges ) of @xmath17 not in the core correspond to identifiable hyperedges ( resp.identifiable vertices ) of @xmath27 .
more information about graph cores can be found in , and hypergraph cores are considered by .
the identifiable vertices obtained by hypergraph collapse also serve to generalize to hypergraphs the definition of graph component .
a graph is a hypergraph having edges only of cardinality two , and consequently has no patches .
however , if the single hyperedge @xmath28 is added to the graph [ making it a hypergraph ] , then the identifiable vertices obtained by running hypergraph collapse on the augmented graph are exactly the vertices in the graph component containing @xmath23 .
the identifiable edges are all the edges of this graph component .
this motivates the following definition for patch - free hypergraphs : a vertex is _ in the domain _ of @xmath23 if it is in the set of identifiable vertices when the hypergraph is augmented by the addition of the hyperedge @xmath28 .
the purpose of this paper is to study the time - evolution of the set of identifiable vertices and the set of identifiable edges in a _ poisson hypergraph process _ , which is a hypergraph - valued , continuous - time stochastic process .
the vertex set is @xmath29 , and the process depends on parameters @xmath30 .
attached to each subset @xmath21 of @xmath0 is a poisson clock run at rate @xmath31 , and these clocks are independent of one another .
[ here @xmath32 denotes the cardinality of @xmath21 .
] when the clock associated to @xmath21 `` rings '' , a new hyperedge equal to @xmath21 is added to the hypergraph . the overall rate at which hyperedges of cardinality @xmath6 are added is then @xmath33
. we will call this process the poisson(@xmath34 ) hypergraph process .
this is a generalization of the ordinary random graph process , in which edges form between each pair of vertices independently at a fixed rate . while for @xmath1 fixed , this process depends only on the finite sequence @xmath35 , we will be interested in the asymptotic behavior as @xmath36 , so we will assume that always the infinite sequence @xmath37 is given .
moreover , this sequence is required to be a probability distribution on @xmath38 with finite expectation and satisfying @xmath7 . the generating function @xmath39
will be denoted by @xmath34 . in , the _
poisson(@xmath40 ) random hypergraph _ is defined , where @xmath41 is a sequence of positive real numbers .
this is a random hypergraph with vertex set v = @xmath42 , so that for each @xmath43 , the number of occurrences of the hyperedge @xmath21 is a poisson random variable with expectation @xmath44 , and these random variables are independent for different subsets of @xmath0 . if @xmath45 is a poisson(@xmath34 ) hypergraph process , then for fixed @xmath46
, @xmath47 is a poisson(@xmath48 ) random hypergraph .
we separate out two distinct cases in the study of poisson hypergraph processes , depending on whether @xmath12 or @xmath49 . when @xmath49 , the hypergraph never acquires patches , and provided the initial hypergraph is patch - free , the set of identifiable vertices is forever void .
as the previous discussion of ordinary graphs suggests , it is natural to consider in such cases the set of vertices in the domain of a distinguished vertex .
we discuss now the case @xmath12 .
our first result describes the evolution of the rescaled number of identifiable vertices and hyperedges in the poisson(@xmath34 ) hypergraph process
let @xmath50 the _ structure function _
@xmath51 , defined as @xmath52 plays a central role for hypergraph processes .
[ recall that @xmath53 . ]
typically @xmath51 is not invertible , but there is a right - continuous monotonic function called the _ lower envelope _ : @xmath54 also important for hypergraph processes is the _ upper envelope _ : @xmath55 we classify structure functions into three types : graph - like , bicritical , and exceptional .
this taxonomy is given in table [ tab : class ] .
figure [ fig : bicritical ] shows a bicritical structure function and the corresponding lower envelope .
.classification of structure functions [ tab : class ] [ cols="^,^,^",options="header " , ] for simplicity , we focus on the case where @xmath56 has a single discontinuity , located at @xmath57 ; i.e. @xmath58 .
the general case follows the same pattern as theorem [ thm : intro_lambda_geq_one ] , because after the number of identifiable vertices has reached @xmath10 , the subsequent evolution is much the same as the @xmath12 case . in the @xmath49 and @xmath59 regime ,
another structure function besides comes into play , namely the structure function @xmath60 of the graph which results from discarding all hyperedges of cardinality more than two : @xmath61 since @xmath60 is monotonic , the corresponding lower envelope @xmath62 defined as @xmath63 is continuous . as before ,
@xmath64 for @xmath65 , and @xmath66 as @xmath67 ; it describes the asymptotic proportion of vertices in the giant component of a random graph where the ratio of edges to vertices is @xmath68 .
we will construct in section [ sec : patch - free_process ] an increasing process @xmath69 so that the distribution of @xmath70 is @xmath71 where @xmath72 is the largest solution @xmath73 in @xmath74 $ ] of @xmath75 . [
notice that @xmath76 for @xmath77 , and @xmath78 otherwise . ]
write @xmath79 for the number of vertices in the domain of @xmath80 in @xmath47 , and write @xmath81 for the number of hyperedges identifiable from @xmath80 in @xmath47 . set @xmath82 and @xmath83 . also , define @xmath84\log(1 - g(t ) ) \right\ } { { \bf 1}}_{\{m_t = \infty\ } } \,.\end{aligned}\ ] ] [ thm : intro_no_patch_process ] consider a poisson hypergraph process such that @xmath85 , and suppose @xmath56 has a single discontinuity located at @xmath57 . fix a distinguished vertex @xmath80 .
the number of vertices in the domain of @xmath80 , and number of hyperedges identifiable from @xmath80 , obey the following limits in distribution as @xmath36 : @xmath86 where we adjoin @xmath87 to @xmath88 as a compactifying point .
also @xmath89 observe the difference between the limit law @xmath90 in and the limit law @xmath91 in : @xmath92 conforms to the deterministic lower envelope @xmath93 , except at points in the finite discontinuity set , whereas @xmath94 waits until the random time @xmath95 , with distribution function @xmath96 , before jumping from @xmath97 up to @xmath93
. see remark [ rmk : no_conv_in_d ] as to whether the convergence extends to weak convergence in the skorohod space @xmath98 .
the rest of this paper is organized as follows : some definitions concerning hypergraphs are given in section [ sec : definitions ] .
we establish that certain key processes are markov in section [ sec : markov ] . the case of hypergraphs and hypergraph processes with patches
are treated in section [ sec : review ] and section [ sec : patches ] respectively .
theorem [ thm : intro_lambda_geq_one ] is proved in section [ sec : patches ] .
patch - free random hypergraphs and hypergraph processes are treated in section [ sec : patch - free_static ] and section [ sec : patch - free_process ] , respectively .
theorem 2 is proved in section [ sec : patch - free_process ] .
finally , we mention future directions in section [ sec : future ] .
recall from the introduction that the identifiable vertices are those vertices removed by the hypergraph collapse algorithm described there , and the identifiable hyperedges are those hyperedges consisting only of identifiable vertices . given a hypergraph @xmath17 and a subset @xmath99
, @xmath100 denotes the hypergraph after all vertices in @xmath101 are deleted .
more precisely , @xmath102 we now more exactly specify the hypergraph collapse algorithm : select if possible a vertex @xmath23 with @xmath103 ; replace @xmath0 by @xmath104 and @xmath17 by @xmath105 ; then repeat .
when the algorithm terminates , we obtain a set @xmath106 consisting of the identifiable vertices , and a patch - free hypergraph @xmath107 on @xmath108 .
suppose @xmath17 is a patch - free hypergraph , and thus having no identifiable vertices .
given such a hypergraph @xmath17 and a distinguished vertex @xmath80 , we say that @xmath23 is in the _ domain _ of @xmath80 in @xmath17 if @xmath23 is identifiable in the hypergraph @xmath109 obtained by augmenting @xmath17 by the hyperedge @xmath110 .
a hyperedge is said to be _ identifiable _ from @xmath80 if it is identifiable in @xmath111 .
_ warning _ : for a general patch - free hypergraph , it is possible for vertex @xmath112 to be in the domain of @xmath23 , while @xmath23 is not in the domain of @xmath112 , although this can not happen in graphs ; see figure [ fig : identifiable ] .
makes @xmath112 identifiable , but not vice versa .
[ fig : identifiable ] ]
for @xmath113 a sequence of non - negative numbers , a _
poisson(@xmath40 ) random hypergraph _ is a random hypergraph @xmath17 with vertex set @xmath114 so that for @xmath43 , the random variable @xmath115 has a poisson distribution with mean @xmath116 , and @xmath117 is a collection of independent random variables . in what follows , @xmath37 will be a probability distribution on the positive integers which has finite mean and @xmath118 we now give an explicit construction of the hypergraph - valued stochastic process described in the introduction .
let @xmath119 be a sequence of independent random variables in @xmath120 with common distribution @xmath121 , for all @xmath122 .
denote by @xmath123 a sequence of independent random subsets of @xmath0 , such that @xmath124 is chosen uniformly at random from the subsets of @xmath0 of size @xmath125 whenever @xmath126 ; the set @xmath124 is not defined when @xmath127 .
let @xmath128 be a poisson process , run at rate @xmath1 , having arrival times @xmath129 .
define a stochastic process @xmath45 with values in the set of hypergraphs with vertex set @xmath0 by @xmath130 interpret @xmath131 as the number of occurrences of hyperedge @xmath21 by time @xmath51 . in summary , for each @xmath43 , @xmath132 and all these poisson processes are independent .
we call @xmath45 a
_ poisson(@xmath34 ) hypergraph process _
, where @xmath34 denotes the generating function @xmath133 the finite mean assumption is equivalent to @xmath134 . for fixed @xmath46 , @xmath47 is a poisson(@xmath135 ) random hypergraph .
whereas the hypergraph literature has tended to concentrate on the `` @xmath6-uniform '' case ( i.e. @xmath136 for some @xmath6 ) , we find the superposition of @xmath6-uniform random hypergraphs for various different values of @xmath6 can be handled without special effort , and leads to asymptotic properties absent from the @xmath6-uniform case .
moreover the poisson structure simplifies our arguments , for example by allowing some summary statistics of @xmath137 to be markov processes in their own right : see proposition [ prop : markov_property ] .
poissonization is , of course , a well - established procedure see .
previous literature has also concentrated on the case @xmath138 .
we now sketch a way to deduce from our results for a poisson(@xmath40 ) random hypergraph @xmath17 some corresponding results for @xmath139 .
we note moreover that if @xmath136 for some @xmath6 then @xmath139 is exactly a @xmath6-uniform hypergraph .
the set of identifiable vertices is the same for @xmath17 and @xmath139 but @xmath17 may have additional identifiable hyperedges .
first consider patches . throwing a poisson(@xmath140 ) number of balls ( i.e. patches ) uniformly at random into @xmath1 urns yields a binomial(@xmath1,@xmath141 ) number of occupied urns ( i.e.vertices covered by at least one patch ) .
hence the number of patches in @xmath17 , less the number in @xmath139 , divided by @xmath1 , has limit in probability @xmath142 . on the other hand ,
the expected number of subsets of size at least 2 receiving at least 2 hyperedges is bounded uniformly in @xmath1 .
hence , after rescaling by @xmath143 , only the extra patches in @xmath17 can contribute in the limit and of course all of these do so .
[ prop : markov_property ] let @xmath144 and @xmath145 denote the numbers of identifiable vertices and identifiable hyperedges for @xmath47 . both @xmath146 and @xmath147 are markov processes .
the number of non - identifiable hyperedges in @xmath47 , given that @xmath148 , is conditionally poisson , with mean @xmath149 \,.\ ] ] when @xmath150 , for @xmath151 $ ] , this reduces as @xmath152 to @xmath153 + o(n ) \,.\ ] ] because the total number of hyperedges in @xmath47 is poisson(@xmath154 ) , proposition [ prop : markov_property ] reduces the study of limits of identifiable hyperedges to study of limits of identifiable vertices . in particular ,
if @xmath155 converges in distribution as @xmath36 to a random variable @xmath156 , then necessarily @xmath157 \,.\ ] ] it is easy to identify the generator of @xmath146 , rescale by division by @xmath1 , and take a limit on any compact interval @xmath158 ( see ) ; however this approach did not lead to a proof of theorem [ thm : intro_lambda_geq_one ] , because of the difficulty of passing through discontinuous phase transitions . to prepare for the proof
, some measure - theoretic apparatus is needed .
let @xmath159 be the probability space on which the process @xmath45 is defined .
for any set @xmath99 , and any @xmath46 , define the @xmath160-field @xmath161 as @xmath162 let @xmath163 denote the set of vertices identifiable at time @xmath51 . by construction ,
the event @xmath164 occurs if and only if , among all sets containing all vertices covered by patches , @xmath101 is the minimal subset of @xmath0 for which @xmath165 whenever @xmath166 . thus @xmath167 .
when we consider @xmath168 as a `` stopping set '' for a set - indexed process , it becomes natural to define another @xmath160-field : @xmath169 @xmath170 and @xmath171 are @xmath172-measurable , for all @xmath173 .
we may describe @xmath172 informally as the knowledge we have about @xmath174 after performing hypergraph collapse at each time @xmath175 $ ] .
[ lem : stopset ] [ it : stopset1 ] fix any @xmath176 .
pick any collection of non - negative integers @xmath177 , and set @xmath178 then @xmath179 [ it : stopset2 ] fix any @xmath176 . the conditional distribution of the random hypergraph @xmath180 ( in the notation of ) , given @xmath172 , on the event @xmath181 , where @xmath182 , is that of a poisson(@xmath40 ) random hypergraph on @xmath183 vertices with parameters @xmath184 for a random variable @xmath185 , we write @xmath186 to indicate that the distribution of @xmath185 is poisson with expectation @xmath187 .
also we will write @xmath188 to indicate that @xmath185 is a binomial random variable with parameters @xmath189 and @xmath190 .
certainly @xmath191 is @xmath172-measurable .
it remains to show that , for any @xmath192 , @xmath193 split the event on the right into disjoint events by intersecting with @xmath181 for each @xmath99 .
for each @xmath101 , @xmath194 lies in @xmath195 , and therefore is independent of @xmath196 for every @xmath21 such that @xmath197 , by construction of a poisson hypergraph process .
the right side becomes @xmath198 which is equal to the left side ; follows .
suppose @xmath99 and @xmath199 with @xmath200 . for any @xmath201 with @xmath202
, implies that @xmath203 the result of part implies that the random variables @xmath204 are conditionally independent for different choices of @xmath205 , given @xmath206 .
if @xmath182 , there are @xmath207 choices of @xmath205 , and following the notation of , @xmath208 in a poisson(@xmath40 ) random hypergraph on @xmath209 vertices , the number of occurrences of @xmath21 , where @xmath210 , is poisson with parameter @xmath211 on comparison with the previous line , this verifies the formula for @xmath212 , when @xmath213 .
clearly there are no @xmath8-hyperedges in @xmath180 when @xmath214 , by definition of identifiability . hence is established .
fix any @xmath176 .
suppose that @xmath148 .
the first jump in the process @xmath215 can occur only when a new hyperedge arrives , and the arrival time is independent of the past .
the law of the jump depends only on two things : the set @xmath21 of vertices in the new hyperedge ( which is independent of the past ) , and on the hypergraph @xmath180 , where @xmath216 .
lemma [ lem : stopset ] establishes that the law of @xmath180 , conditional on @xmath172 is fully determined by @xmath217 , @xmath51 , and the parameters @xmath218 ; in particular it is conditionally independent of @xmath219 given that @xmath220 .
hence the markovian property of @xmath146 and @xmath147 is established .
it follows from lemma [ lem : stopset ] that the total number of non - identifiable hyperedges in @xmath47 , given that @xmath220 , is conditionally poisson , with mean @xmath221 , for @xmath212 as in .
write @xmath222 , and switch the order of summation , to obtain @xmath223 on considering the hypergeometric(@xmath224 ) distribution , we see that the inner sum is @xmath225 / \binom{n}{k } \,.\ ] ] the last expression is zero when @xmath226 , so @xmath227 takes the form .
when @xmath228 , the last expression converges , as @xmath36 , to @xmath229 , and is bounded between @xmath97 and @xmath8 .
the bounded convergence theorem yields .
in this section we review some material from .
fix @xmath176 , and set @xmath230 , @xmath231 . in this case , @xmath17 is a poisson(@xmath40 ) random hypergraph .
suppose we perform hypergraph collapse , described above , in the following special way : at each step the next vertex @xmath23 to be deleted is selected with a probability proportional to the number of patches on @xmath23 .
this is called _ randomized collapse_. the _ debris _ of a hypergraph is the number of hyperedges equal to the empty set . set @xmath232 , and let @xmath233 denote the sequence of hypergraphs obtained . set @xmath234 and @xmath235 to be the amount of patches and debris , respectively , in @xmath236 ; formally @xmath237 the key observation in is that @xmath238 is a markov chain ( but not the same one as in proposition [ prop : markov_property ] , for here @xmath51 is fixed ! ) , which stops at @xmath239 moreover , conditional on @xmath240 , @xmath241 here @xmath242 and @xmath243 are independent , with @xmath244 where @xmath245 by construction , @xmath246 , the number of identifiable vertices , and @xmath247 is the number of identifiable hyperedges . for comparison ,
note that , by proposition [ prop : markov_property ] the number of non - identifiable hyperedges in @xmath17 , given that @xmath248 , is conditionally poisson , with mean @xmath249 we obtained a limit theorem for @xmath250 and @xmath251 , where @xmath252 is the number of identifiable hyperedges .
we state the result in a simple case .
set @xmath253 \,.\ ] ] assume that @xmath254 and that the derivative @xmath255 .
then @xmath256 is non - empty , and its infimum is @xmath93 , as defined in . by our assumption , there is at most one @xmath257 such that @xmath258 , namely @xmath259 ; this is different to @xmath93 only if @xmath260 , the set of discontinuity points of the lower envelope @xmath56 .
let @xmath261 be a random variable taking values @xmath93 and @xmath259 , each with probability @xmath262 .
as a special case of of * theorem 2.2 we know : [ thm : fixedtime ] the following limit in distribution holds as @xmath36 : @xmath263 have shown that the limit for the rescaled number of identifiable hyperedges can be decomposed as follows : @xmath264 counts the _ essential hyperedges _ ,
i.e. those whose absence would have reduced the set of identifiable vertices , and @xmath265 counts the remainder .
suppose in particular that @xmath230 and @xmath266 for some @xmath260 , the discontinuity set of @xmath56 .
then implies that the proportion of identifiable vertices has a limit in distribution which is random , taking the values @xmath93 and @xmath259 each with probability @xmath262 .
it suffices to derive the limit for @xmath267 , since the limit for @xmath268 follows from proposition [ prop : markov_property ] . to check this , recall
that , by , if @xmath267 converges to @xmath93 , then the number of identifiable hyperedges , divided by @xmath1 , converges to @xmath269\rho ' ( g(t ) ) \right\ } \,.\ ] ] however by definition of @xmath93 , @xmath270 , so we have recovered the formula @xmath271 .
in this section we move from the static random hypergraph model of theorem [ thm : fixedtime ] to the poisson(@xmath34 ) hypergraph process @xmath45 , providing here a proof of theorem [ thm : intro_lambda_geq_one ] . extending the notation of the previous section ,
let @xmath272 and @xmath273 denote the rescaled numbers of identifiable vertices and hyperedges for @xmath47 , respectively , as defined in .
note that @xmath274 and @xmath275 are increasing , right - continuous , stochastic processes .
it follows from proposition [ prop : markov_property ] that @xmath276 is a markov process .
fix @xmath277 .
we have to show the convergence in distribution @xmath278 it suffices to do so when at least one of @xmath279 is not a discontinuity point , for every @xmath280 .
proposition [ prop : markov_property ] showed that @xmath281 is markov , and for any markov process @xmath282 the conditional law of @xmath283 given @xmath284 is the same as the conditional law given @xmath285 .
hence it suffices to consider the case @xmath286 such that @xmath287 or @xmath288 , and these possibilities are both subsumed in the case @xmath289 with @xmath290 .
then only the marginal limit at time @xmath60 , as given in theorem [ thm : fixedtime ] is random , so theorem [ thm : fixedtime ] implies the full convergence in distribution .
the second assertion follows from the first since all processes are increasing , and the limit is deterministic and continuous on @xmath18 .
[ rmk : no_conv_in_d ] the rescaled number of essential hyperedges , as studied by , has a limit @xmath291 in the same sense as and .
one may ask whether the convergence extends to weak convergence in the skorohod space @xmath292 .
since @xmath293 and @xmath275 are non - decreasing , the necessary and sufficient condition of * p. 306 may be applied , which would require that the sum of squared jumps of @xmath294 converges in law to the sum of squared jumps of @xmath295 , and similarly for @xmath296 .
unfortunately the techniques presented in this paper do not seem to be able to confirm this ; indeed , it seems plausible that , for arbitrarily large @xmath1 , and for @xmath260 , there is a probability bounded away from zero that @xmath297 makes more than one jump in going from @xmath298 to @xmath299 at time @xmath300 , and this would contradict the condition stated .
if is false , one can reformulate the process , by consulting * theorem 2.2 and prove a corresponding version of theorem [ thm : intro_lambda_geq_one ] .
we revert to the fixed - time setting of section [ sec : review ] .
suppose @xmath17 is a poisson(@xmath40 ) random hypergraph , such that @xmath301\,.\ ] ] fix a vertex @xmath80 .
write @xmath302 for the number of vertices in the domain of @xmath80 , and write @xmath303 for the number of hyperedges identifiable from @xmath80 . set @xmath304 and @xmath305 . both the _ microscopic variables _ @xmath306 , and the _ macroscopic variables _ @xmath307 have non - trivial limits as @xmath308 , which we now describe .
the coefficient @xmath309 plays a distinguished role .
let @xmath310 be a random walk on the integers , started at @xmath311 , whose increments are of the form @xmath312 .
let @xmath313 be the largest root in @xmath74 $ ] of @xmath314 , so @xmath315 for @xmath316 , and @xmath317 otherwise .
then the first passage time to @xmath97 , @xmath318 has the following distribution : @xmath319 @xmath320 is distributed as the total number of individuals in a branching process with one ancestor , and poisson(@xmath321 ) offspring distribution .
this distribution describes the sizes of small components in an erds - rnyi random graph ; see .
the fact that @xmath322 is an elementary fact from the theory of branching processes .
the formula for @xmath323 is a special case of a formula of , which is proved in detail on p. 300 of .
assume that @xmath324 .
then the set is non - empty , and its infimum is @xmath325 , as defined in .
assume further that @xmath326 for all @xmath327 .
if either of these assumptions fail , then the techniques of , combined with some arguments given below , still establish the desired asymptotics .
we omit the details . set @xmath328 { { \bf 1}}_{\{m = \infty\}}\ , .
\end{split}\ ] ] [ thm : identif_sans_patches ] consider a poisson random hypergraph without patches , and fix a distinguished vertex @xmath80 .
the number of vertices in the domain of @xmath80 , and number of hyperedges identifiable from @xmath80 , obey the following limits in distribution as @xmath36 : @xmath329 here @xmath320 is considered as a random variable taking values in the one - point compactification @xmath330 of @xmath88 .
_ step i_. set @xmath331 , and let @xmath233 be a sequence of hypergraphs obtained by randomized collapse
. denote by @xmath332 and @xmath333 the numbers of patches and debris , respectively , in @xmath236 .
then @xmath334 we know that @xmath335 is a markov chain , starting from @xmath336 : the increments , conditional on @xmath337 and @xmath338 , are as given in and . for fixed @xmath339 and @xmath340 , the random variable @xmath242 defined in converges to @xmath97 in distribution as @xmath36 .
also @xmath341 so the random variable @xmath243 defined in converges to poisson(@xmath321 ) in distribution as @xmath36 . hence , for all @xmath339 , @xmath342 which implies @xmath343 as @xmath36
if @xmath344 , then @xmath345 , so the proof is complete . it only remains to prove the second convergence assertion in the case where @xmath346 , and @xmath317 .
_ step ii_. introduce an auxiliary time variable @xmath51 , and let @xmath347 be a poisson process of rate @xmath1 . set @xmath348 with reference to , set @xmath349 by theorem 6.1 and remark 6.2 of , for all @xmath350 , @xmath351
observe that @xmath352 , which will have the same limit in probability as does @xmath353 .
we will show that , for all @xmath354,0)$ ] , there exists @xmath350 and @xmath355 such that @xmath356 by and the fact that @xmath357 , we know that , for all @xmath350 and all @xmath358 : @xmath359 for all sufficiently large @xmath1 . also from we obtain , for all @xmath350 , @xmath360 as @xmath36 .
hence the claim that @xmath361 will follow as soon as we have proved ; then will strengthen this to show @xmath362 .
_ step iii_. the remainder of the proof is to establish . given @xmath337 , set @xmath363 where @xmath364 lemma 6.1 of implies that @xmath365 as @xmath36 .
since @xmath366 , there is @xmath367 such that @xmath368 ; by construction of @xmath313 , @xmath369 , so @xmath370 ; in other words , @xmath371 we can therefore find @xmath350 and @xmath355 such that @xmath372 consider the martingale @xmath373 and set @xmath374 .
it follows from that , on the event @xmath375 , @xmath376 hence for @xmath377 , @xmath378 however implies that , for @xmath379 , @xmath380 , and follows .
we now focus on the case of patch - free hypergraph processes , proving in this section theorem [ thm : intro_no_patch_process ] .
let @xmath381 , @xmath382 , be a family of independent poisson processes , all of rate @xmath383 , and consider the coupled family of random walks @xmath384 , for @xmath385 , where @xmath386 for all @xmath189 , and @xmath387 the marginal law of @xmath70 is given by with @xmath388 .
there is a relation between @xmath384 and the multigraph structure function : since @xmath96 is the largest root in @xmath74 $ ] of @xmath389 , we have as a special case of : the first time @xmath51 at which @xmath384 escapes to infinity is related to the multigraph lower envelope as follows : @xmath390 moreover @xmath391 is an increasing process by the coupling , so @xmath392 is a continuous random variable with distribution function @xmath96 .
we finally turn to the case of a poisson(@xmath34 ) hypergraph process @xmath45 without patches , i.e. such that @xmath393 \,.\ ] ] write @xmath79 for the number of vertices in the domain of @xmath80 in @xmath47 , and write @xmath81 for the number of hyperedges identifiable from @xmath80 in @xmath47 . set @xmath82 and @xmath83 . using , we define what will turn out to be the macroscopic limits for theorem [ thm : intro_no_patch_process ] .
@xmath84\log(1 - g(t ) ) \right\ } { { \bf 1}}_{\{m_t = \infty\ } } \,.\end{aligned}\ ] ] _ step i_. extending the notation of theorem [ thm : identif_sans_patches ] let @xmath394 denote the hypergraph that results from applying @xmath189 steps of randomized collapse to @xmath395 ; @xmath396 and @xmath397 count the number of patches , and the amount of debris , respectively in @xmath394 , and @xmath189 is assumed to satisfy : @xmath398 consider a finite set of time points @xmath399 .
the hypergraph collapses of @xmath400 are coupled together as follows : perform the @xmath401st step of randomized collapse by choosing a patch uniformly at random from the smallest unstable hypergraph .
poisson symmetries imply that this amounts to randomized collapse for each of the unstable hypergraphs .
condition on the event : @xmath402 for @xmath403 such that @xmath404 , evidently @xmath405 and @xmath406 . for those @xmath403 such that @xmath407 ,
we may write : @xmath408 where the random increments are distributed as follows .
take @xmath409 to be the least @xmath410 for which @xmath411 , and take @xmath412 and @xmath413 independent such that @xmath414 where @xmath415 is as in .
because of the coupling , we may take subsequent increments ( for @xmath416 ) to be independent and of the form : @xmath417 _ step ii_. observe that the behavior of @xmath415 depends on whether @xmath418 , or @xmath419 .
it follows from and the calculations in step i that , conditional on , the joint law of @xmath420 converges as @xmath36 to the conditional law of @xmath421 given that @xmath422 .
evidently @xmath423 for all @xmath403 . since @xmath189 was arbitrary , and
since for each @xmath51 both @xmath384 and @xmath424 are markov , we have now proved convergence in distribution as @xmath36 : @xmath425 in particular , in the notation of and section [ sec : no_patch ] , @xmath426 _ step iii_. to prove it suffices , in the light of , to prove tightness of @xmath427 with respect to the skorohod topology of @xmath428 .
on @xmath429 , we shall use the metric @xmath430 understanding that @xmath431 .
we shall verify the condition of aldous for tightness of @xmath432 , as stated in , p. 176 , or , p. 314
, with respect to this metric .
since @xmath433 and @xmath434 are non - decreasing processes , the condition takes a slightly simpler form than usual : it suffices to show that , for each @xmath435 and @xmath436 , there exist @xmath437 and @xmath355 such that for every bounded sequence of optional times @xmath438 with respect to @xmath427 , and for every @xmath439 , @xmath440 where @xmath160 is short for @xmath438 in the subscripts .
proposition [ prop : markov_property ] established that @xmath441 is a markov process . by the strong markov property , the conditional law of @xmath442 , given that @xmath443 , and @xmath444 , is that same as that of the number of identifiable vertices in a poisson(@xmath445 ) random hypergraph @xmath446 on @xmath447 vertices , where by the reasoning of lemma [ lem : stopset ] and the fact that @xmath49 , @xmath448 suppose @xmath435 and @xmath436 are given . in the case where @xmath449 , it follows that @xmath450 on the other hand , if @xmath451 , then @xmath452 choose @xmath355 so large that , for @xmath439 , the right side is not more than @xmath453 ; now it is true that , for any @xmath454 and for any @xmath439 , the probability that @xmath446 has no patches , and hence no identifiable vertices nor identifiable hyperedges , is at least @xmath455 ; in that case , @xmath456 and and @xmath457 .
in summary , for such @xmath1 and @xmath437 , holds . hence @xmath427 is tight , and follows .
_ step iv_. as for we need only check the convergence of finite - dimensional distributions , i.e. that @xmath458 for every finite set of time points @xmath399 . for the case
@xmath459 , the validity of follows from theorem [ thm : identif_sans_patches ] . for the sake of brevity , restrict our discussion of the case @xmath460 to the @xmath461 component ;
the argument for the @xmath462 component is similar .
it suffices to show , for all @xmath463 , and all @xmath435 , that @xmath464 by our knowledge of the finite dimensional distributions from theorem [ thm : identif_sans_patches ] , the left side of is well approximated by @xmath465 and for @xmath466 sufficiently small , this converges to the right side of .
we have not explained here the role of the upper envelope , even though it was included in the classification of structure functions .
it is related to dual hypergraph collapse and the size of the core , as in
. we shall give the corresponding asymptotic results in a future paper .
we thank peter matthews and the referees for suggesting various expository improvements . | let @xmath0 denote a set of @xmath1 vertices . to construct a _ hypergraph process _
, create a new hyperedge at each event time of a poisson process ; the cardinality @xmath2 of this hyperedge is random , with generating function @xmath3 , where @xmath4 ; given @xmath5 , the @xmath6 vertices appearing in the new hyperedge are selected uniformly at random from @xmath0 .
assume @xmath7 .
hyperedges of cardinality @xmath8 are called _ patches _ , and serve as a way of selecting root vertices .
identifiable vertices are those which are reachable from these root vertices , in a strong sense which generalizes the notion of graph component .
hyperedges are called identifiable if all of their vertices are identifiable .
we use `` fluid limit '' scaling : hyperedges arrive at rate @xmath1 , and we study structures of size @xmath9 and @xmath10 .
after division by @xmath1 , numbers of identifiable vertices and hyperedges exhibit phase transitions , which may be continuous or discontinuous depending on the shape of the structure function @xmath11 .
both the case @xmath12 , and the case @xmath13 are considered ; for the latter , a single extraneous patch is added to mark the root vertex .
_ national security agency , university of utah , university of cambridge _ |
the modelling and interpretation of the iso - sws ( infrared space observatory - short wavelength spectrometer ) data require an accurate calibration of the spectrometers ( schaeidt et al .
1996 ) . in the sws spectral region
( 2.38 45.2 @xmath19 m ) the primary standard calibration candles are bright , mostly cool , stars .
the better the behaviour of these calibration sources in the infrared is known , the more accurate the spectrometers can be calibrated .
iso offered the first opportunity to obtain continuous mid - infrared spectra between 2.38 and 45.2 @xmath19 m at a spectral resolving power of @xmath20 , not polluted by any molecular absorptions of the earth s atmosphere .
therefore our knowledge on the mid - infrared behaviour of the stellar calibration sources is limited .
refining the synthetic reference spectra used to calibrate the sws can only be done by refining the model atmospheres of the stars .
a full exploitation of the iso data can therefore only result from an iterative process in which both accurate observations and new modelling are involved . in order to obtain a reliable convergence in such an iterative process
, one needs : ( 1 ) a sample of bright stellar objects spread over a large range in spectral type ; ( 2 ) a thorough understanding of cool star atmospheres and of the influence of the various parameters e.g. t@xmath1 , @xmath3 g , and chemical composition on the emergent spectra . with this in mind
, stars with spectral types ranging from a0 m8 were observed with iso - sws .
it is important to cover a broad parameter space in order to distinguish calibration problems from problems related to the model and/or in the generation of the synthetic spectrum .
concerning point 2 above , it is necessary to perform an intensive study of the influence of different stellar parameters on synthetic spectra .
the results of such a study will be presented with the red giant @xmath0 tau as a test case .
the results for 15 other stars in our sample will be presented in forthcoming papers .
this research has been done within the framework of the proposal starmod ( title : accurate modelling of cool - star atmospheres ; p.i . : c. waelkens ; c.i .
: m. cohen , l. decin , th
. de graauw , l.b.f.m . waters ) and the zzstarm proposal ( title : accurate modelling of cool - star atmospheres ; p.i . : l. decin ; c.i . :
m. cohen , c. waelkens , th .
de graauw , l.b.f.m .
some calibration data have been provided by the sws instrument dedicated team ( sidt ) in the framework of a quick - feedback refining of the model spectral energy distribution ( sed ) of the calibration sources used for the sws calibration .
so far , the analysis of the discrepancies between the iso - sws data and the corresponding synthetic spectra has been restricted to the wavelength region from 2.38 to 12 @xmath19 m , since the lack of comprehensive molecular and atomic line lists hamper fast progress at longer wavelengths ( 12 45 @xmath19 m ) .
furthermore the brightness of the stars drops quickly in this wavelength region so that the same signal to noise ratio will not be achieved .
a third point is that the seds may also be affected by unknown circumstellar contribution .
this paper is organized as follows : in sect .
2 the observations are described and the data reduction procedures are discussed . in sect .
3 a summary of the literature concerning @xmath0 tau is presented , on the basis of which the starting values for our modelling stellar parameters are selected .
the effect of changing stellar parameters on the synthetic spectra of k and m giants is analyzed in sect .
4 . in sect .
5 the method of analysis is outlined , while in sect .
6 the results are discussed . in the last section ,
7 , the main conclusions are summarized .
a full scan observation of @xmath0 tau has been obtained in the 2.38 45.2 @xmath19 m wavelength range with the short - wavelength spectrometer ( sws ) ( de graauw et al .
1996 ) on board iso .
this has been done by use of the sws observing mode aot01 (= a single up - down scan for each aperture / order combination ) with scanner speed 4 , resulting in a resolving power of @xmath21 1500 . the observation lasted for 6650 s and was performed during revolution 636 .
the data were processed to a calibrated spectrum sampled in all the pixels , the so - called auto - analysis product , using the procedures and calibration files of the iso off - line pipeline version 7.0 . for a description of the flux and wavelength calibration ,
we refer to schaeidt et al .
( 1996 ) and valentijn et al .
this reduction resulted in an oversampled spectrum for each of the four aot - bands , which each consist of 12 detectors .
remaining instrumental effects , such as fringes in the 12.0 29.5 @xmath19 m part of the spectrum , were removed using the sws interactive analysis package ( ia ) provided by the sidt .
the band-2 data ( 4.08 12 @xmath19 m ) are severely affected by detector memory effects .
time and flux dependent memory effects occur for all four instruments ( sws , lws , phot , cam ) on board iso .
so far , several attempts have been made to model the corrections of the transient effects ( see e.g.van malderen et al .
1999 ) , but currently no foolproof method exists to correct the sws data for these transient effects .
memory effects appear to be less severe in the down - scan measurements , which are obtained after the up - scan data , suggesting a more stabilized response to the flux level for down - scan data .
the sws flux calibration of band 2 was corrected for a regime where the flux history of the detectors is comparable to the second scan in the observation , the so - called down scan .
we therefore used the down - scan data of our observation as a reference to do a correction of the flux level of the first scan .
several procedures were performed on each sub - band separately to combine the twelve detectors . a sigma - clipping procedure ( with @xmath22 ) was used to reduce the noise by discarding the datapoints in every resolution element for which the difference to the mean was @xmath23 . after aligning the twelve different detector signals to their average level and rebinning to the expected resolution ( see table [ resolution ] ) , the final result was obtained .
although the resolution changes in within one sub - band , the resolution was kept constant ( see table [ resolution ] ) and the value was taken to be the lowest value in fig . 6 by lorente ( 1998 ) .
the individual sub - band spectra , when combined into a single spectrum , can show jumps in flux levels at the band edges .
this is due to imperfect flux calibration or wrong dark - current subtraction for low - flux observations . using the overlap regions of the different sub - bands and looking at other sws observations ,
several sub - bands were multiplied by a small factor ( see table [ resolution ] ) to construct a smooth spectrum . due to the problems with memory effects in band 2 ,
the factor of band 2b was determined by use of the template of cohen et al .
( 1992 ) and the obtained synthetic spectra .
note that all shifts are well within the photometric absolute calibration uncertainties claimed by schaeidt et al .
( 1996 ) . ' '' '' & wavelength & & + ' '' '' [ 0pt]sub - band & range [ @xmath19 m ] & [ 0pt]resolution & [ 0pt]factor + ' '' '' 1a & 2.38 2.60 & 1300 & 1.00 + 1b & 2.60 3.02 & 1200 & 1.01 + 1d & 3.02 3.52 & 1500 & 1.00 + ' '' '' 1e & 3.52 4.08 & 1000 & 1.00 + ' '' '' 2a & 4.08 5.30 & 1200 & 1.00 + 2b & 5.30 7.00 & 800 & 1.045 + ' '' '' 2c & 7.00 12.0 & 800 & 1.00 +
' '' '' t@xmath1 & @xmath3 g & m & @xmath8 & [ fe / h ] & @xmath13(c ) & @xmath13(n ) & @xmath13(o ) & @xmath24c/@xmath25c & @xmath17 & l & r & ref . + ' '' '' @xmath26 & @xmath27 & & @xmath28 & @xmath29 & & & & & & & & 1 . + @xmath30 & @xmath31 & & @xmath32 & @xmath33 & @xmath34 & @xmath35 & @xmath36 & @xmath37 & & 200 & & 2 .
+ & & & & & & & & & 21.154 & & & 3 . + @xmath38 & @xmath39 & & @xmath40 & & @xmath41 & & & & & & & 4 .
+ @xmath38 & @xmath39 & & @xmath40 & & @xmath42 & & & & & & & 4 .
+ @xmath38 & @xmath39 & & @xmath43 & & @xmath44 & & & & & & & 5 .
+ @xmath45 & @xmath46 & 1.5 & @xmath47 & @xmath48 & @xmath49 & @xmath50 & @xmath51 & 10 & & & & 6 .
+ 3831 & 1.41 & 1.5 & 2.0 & & 8.37 & & & @xmath37 & & & & 7 .
+ @xmath52 & @xmath53 & 1.4 & 2.1 & @xmath54 & & & & @xmath55 & _ 24 _ & 413 [ @xmath56 & @xmath57 & 8 .
+ 3790 & 1.8 & & & & & & & 12 & & & & 9 .
+ 3898 & 2.0 & & & & & & & & @xmath58 & & & 10 .
+ 3800 & 1.8 & & 1.0 & @xmath59 & & & & & & & & 11 .
+ @xmath60 & @xmath61 & @xmath62 & 1.9 & 0.00 & & & & & _ 23 _ & @xmath63 & & 12 .
+ 3789 & 1.8 & & 1.5 & & & & & @xmath64 & & & & 13 .
+ @xmath65 & 1.55 & 1.0 & @xmath66 & @xmath67 & @xmath68 & @xmath69 & @xmath70 & @xmath71 & & & & 14 .
+ @xmath65 & 0.55 & 1.0 & @xmath66 & @xmath72 & @xmath73 & @xmath74 & @xmath75 & @xmath71 & & & & 14 .
+ @xmath38 & @xmath39 & & @xmath76 & & @xmath41 & @xmath77 & & & & & & 15 .
+ @xmath78 & @xmath79 & & @xmath80 & @xmath81 & & & & & & & & 16 .
+ 3950 & 1.5 [ 1.3 ] & & 1.5 [ 2.0 ] & 0.00 & & & @xmath82 [ 9.03 ] & & & & & 16 .
+ 3850 & 0.9 [ 0.6 ] & & 1.5 [ 2.0 ] & @xmath83 & & & 8.47 [ 8.53 ] & & & & & 16 .
+ 3750 & 0.5 [ 0.2 ] & & 1.5 [ 2.0 ] & @xmath83 & & & 8.27 [ 8.37 ] & & & & & 16 .
+ & & & & & & & & & @xmath84 & & & 17 .
+ @xmath85 & & & & & & & & & _ @xmath86 _ & & & 18a .
+ @xmath87 & & & & & & & & & _ 21.07 _ & & & 18b . + & & & & & & & & & @xmath88 & & & 19 .
+ & & & & & & & & & @xmath89 & & & 20 .
+ & & & & & & & & & @xmath90 & & & 20 .
+ @xmath91 & & & & & & & & & & @xmath92 & @xmath93 & 21 .
+ @xmath94 & & & & & & & & & 20.634 & & & 22 .
+ 3690 & & & & & & & & & & & & 23 .
+ 3943 & 1.2 & & & @xmath95 & & & & & _ 20.62 _ & & & 24 .
+ & & & & @xmath96 & & & & & & & & 25 .
+ @xmath97 & @xmath98 & & & @xmath99 & & & & & & & & 26 .
+ & & & & & & & & & 20.035.0 & & 23.0 & 27@xmath100 . + & & & & & & & & & _ 16.024.0 _ & & 45.052.0 & 27@xmath101 .
+ @xmath102 & & & & & & & & & @xmath103 & & & 28 .
+ 3840 & & & & & & & & & 21.4 & & & 29 .
+ 3820 & & & & & & & & & @xmath104 & & & 30 .
+ @xmath105 & & & & & & & & & @xmath106 & & & 31 .
+ 4000 & & & & & & & & & @xmath107 & & & 32 .
+ ' '' '' @xmath108 & @xmath46 & 1.5 & @xmath47 & @xmath48 & @xmath109 & @xmath110 & @xmath111 & @xmath112 & & & & 33 .
+ @xmath0 tau (= hd 29139 = hr 1457 = hic 21421 ) is classified as a k5iii giant .
a detailed literature study was necessary , on the one hand to extract the best possible set of starting fundamental parameters in order to reduce the number of calculated spectra and on the other hand to check the consistency between the stellar parameters deduced from the iso - sws data and other spectra / methods .
an exhaustive discussion of published results is presented in table [ literature ] and in appendix a. after evaluation of the various sources , the stellar parameters of mcwilliam ( 1990 ) , with t@xmath1 = 3900 k , @xmath3 g = 1.60 , @xmath8 = 2.0 @xmath10 , [ fe / h ] = @xmath113 , together with @xmath13(c ) = 8.27 , @xmath13(n ) = 7.86 , @xmath13(o ) = 8.60 ( lambert & ries ( 1981 ) , but with [ n / fe ] taken to be @xmath114 according to luck & challener ( 1995 ) , see appendix a ) , a mass of 1.5 m@xmath6 ( van paradijs & meurs 1974 ) , the limb - darkened angular diameter @xmath17 @xmath115 mas from ridgway et al .
( 1982 ) and @xmath11 @xmath116 ( smith & lambert 1985 ) were taken as starting values . in fig .
[ startvalues ] both band 1 of the iso - sws data of @xmath0 tau and the synthetic spectrum generated with these parameters are shown . from fig .
[ startvalues ] , it is already obvious that not only the angular diameter @xmath17 should be lowered and/or t@xmath1 should be increased , but also other parameters may need improvement . [ 0.4 ]
in this section the effect of stellar parameters on the absorption by co , sio , oh and h@xmath117o is discussed .
we mainly focus on these molecules since they are the most prominent absorbers for oxygen - rich giants in the wavelength range considered here ( 2.38 12 @xmath19 m ) ( see , e.g. , fig . 2 and fig . 3 in decin et al .
( 1997 ) and fig .
[ atdiv ] in this paper ) .
the goal of this part of the study is to learn how the synthetic spectrum will change when one of the parameters is changed within its uncertainties , the only exception being the stellar mass and the @xmath11 ratio due to their small influence .
a simple equation for any chemical compound is not easily obtained , nor is a unique scenario able to explain all different behaviours , but some patterns do emerge .
these results are then useful for the determination of the origin of the discrepancies seen between iso - sws data and synthetic spectra .
since the strength of molecular ( and atomic ) lines is proportional to the ratio of line to continuous absorption coefficient , @xmath118 , approximations for this ratio are sought for .
the approximate ratio is referred to as @xmath119 and equations for some molecules have been deduced according to the approach of kjgaard et al .
( 1982 ) ( see appendix b ) .
the ratio @xmath119 has been studied for the surface flux using the following parameters : + t@xmath1 = 3650 k 3850 k
4050 k + @xmath3 g = 1.00 1.50 2.00 + z = @xmath120 0.00 0.15 + m = 1.5 m@xmath6 10 m@xmath6 15 m@xmath6 + @xmath8 = 1.0 @xmath10 2.0 @xmath10 + @xmath11 = 10 4.26 + @xmath13(c ) = 8.24 8.54 + @xmath13(n ) = 8.26 8.56 + @xmath13(o ) = 8.83 9.13 + all computed spectra are compared with the synthetic spectrum with parameters : t@xmath1 = 3650 k , @xmath3 g = 1.00 , z = 0.00 , m = 1.5 m@xmath6 , @xmath8 = 2.0 @xmath10 , @xmath11= 10 , @xmath13(c ) = 8.24 , @xmath13(n ) = 8.26 , @xmath13(o ) = 8.83 . if not specified these stellar parameters are used . in fig .
[ overall ] the overall results of a change @xmath121t@xmath1 = 200 k ( a ) , @xmath122 g = 0.50 ( b ) , @xmath121 m = 13.5 m@xmath6 ( c ) , @xmath121z = 0.15 ( d ) , @xmath121@xmath8 = 1.0 @xmath10 ( e ) , @xmath121@xmath11 = 5.84 ( f ) , @xmath123(c ) = 0.30 ( g ) , @xmath123(n ) = 0.30 ( h ) , @xmath123(o ) = 0.30 ( i ) are shown .
the results mentioned and discussed in this study apply only to models with similar stellar parameters . [
.8 ] the models and corresponding synthetic spectra have been computed by using the marcs - code ( gustafsson et al .
1975 ) . since 1975
, this code has undergone some modifications , the most important ones being the replacement of the opacity distribution function ( odf ) technique by the opacity sampling ( os ) technique , the possibility to use a spherically symmetric geometry for extended objects and major improvements of the line and continuous opacities ( plez et al . 1992 ) .
the common assumptions of spherical or plane - parallel stratification in homogeneous stationary layers , hydrostatic equilibrium and lte were made .
energy conservation was required for radiative and convective flux , where the energy transport due to convection was treated through a local mixing - length theory ( henyey et al .
the mixing - length @xmath124 was chosen as 1.5 @xmath125 , with @xmath125 the pressure scale height , which is a reasonable quantity to simulate the temperature structure beneath the photosphere ( nordlund & dravins 1990 ) .
turbulent pressure was neglected .
the reliability of these assumptions is discussed by plez et al .
( 1992 ) .
the synthetic spectra were generated using the turbospectrum program described by plez et al .
( 1993 ) , and further updated .
the program treats the chemical equilibrium for hundreds of molecules with a consistent set of partition functions and dissociation energies .
solar abundances from anders & grevesse ( 1989 ) have been assumed , except for the iron abundance , @xmath13(fe)@xmath126 , which is in better agreement with the meteoritic value .
the continuous opacity sources considered here are h@xmath127 , h , fe , ( h+h ) , h@xmath128 , h@xmath129 , he i , he i@xmath130 , he@xmath127 , c i , c ii , c i@xmath130 , c ii@xmath130 , c@xmath127 , n i , n ii , n@xmath127 , o i , o ii , o@xmath127 , co@xmath127 , h@xmath117o@xmath127 , mg i , mg ii , al i , al ii , si i , si ii , ca i , ca ii , h@xmath117(pr ) , he(pr ) , e@xmath131 , h@xmath131 , h@xmath132 , where ` @xmath133 ' stands for ` pressure induced ' and ` sc ' for ` scattering ' . for the line opacity in the sws range ( 2.38 12 @xmath19 m )
a database of infrared lines including atoms and molecules has been prepared . for the atomic lines
the data listed by hirata & horaguchi ( 1995 ) have been used , for co those by goorvitch ( 1994 ) , for sio those by langhoff & bauschlicher ( 1993 ) , for h@xmath117o those by jrgensen ( 1994 ) and ames ( partridge & schwenke 1997 ) , oh lines by sauval ( melen et al .
1995 ) , schwenke ( 1997 ) and goldman et al .
( 1998 ) , nh lines by sauval ( grevesse et al . 1990 ; geller et al . 1991 ) and ch lines by sauval ( melen et al . 1989 ; grevesse et al . 1991 ) and cn lines by plez ( private communication ) .
the dissociation energy for cn was taken to be 7.76 ev .
an exhaustive discussion on the accuracy and completeness of infrared spectroscopic line lists can be found in decin ( 2000 ) . from this study ,
a preference emerged for the h@xmath117o line list by ames and the oh line list by goldman . to remove the small difference between the originally ` vacuum ' wavelengths for the iso observations and ` air ' wavelengths in spectroscopic linelists above 200 nm , edlen s formula ( edlen 1966 )
was extended to the infrared .
when the temperature increases , fewer molecules are formed resulting in a smaller line absorption coefficient . at 2.3 @xmath19 m , @xmath134 is primarily due to h@xmath127 free - free absorption , thus its value depends directly on the electron pressure @xmath135 .
increasing the temperature causes more ionization events and thus more free electrons , but at the same time the h@xmath127 ion itself is less easily formed .
shows that , for the models under consideration , this latter effect is the most important one and so @xmath134 decreases when the effective temperature is increased from 3650 k to 3850 k. moving inwards from the outer photosphere , the line absorption coefficient l@xmath136 of all molecules ( not h@xmath127 ) reaches its maximum at the location where the effect of an increasing temperature overtakes the effect of a higher density .
the partial pressure p of h@xmath127 , being proportional to @xmath137(h i)*@xmath135 , keeps on rising because of the increase of the number of free electrons at higher temperature .
when the line - to - continuum contrast has to be known , one has to consider the partial pressure of the molecule divided by ( @xmath137(h i)*@xmath135 ) ( see fig .
[ overall ] and fig .
[ part]b ) . the simple statement that there are weaker absorption bands for higher temperatures does not always hold : the approximate ratio @xmath119 of co , sio , oh and h@xmath117o decreases , but for cn the lower continuous opacity compensates for the lower line opacity ( see fig .
[ part]b ) .
furthermore , the relative population of rotational levels strongly depends on t ( the maximum population occurs at a j@xmath138-value and a lower t corresponds to a lower j@xmath138-value ) .
therefore a given line ( j ) intensity could be increased or decreased according to its j - value only .
[ 0.35 ] although g and m are closely related , the effect of changing the mass is smaller than that of changing the gravity , because the first one only affects the extension and the latter one also changes the pressure structure of the atmosphere .
this is clearly visible in fig .
[ overall ] .
the assumption of hydrostatic equilibrium links the gas pressure to the surface gravity . in a cool star ,
the gas and electron pressure ( @xmath139 and @xmath135 ) can be written as : @xmath140 due to the higher extension of the atmospheres considered here , this value of @xmath137 has a larger range compared to the value discussed by gray ( 1992 ) .
when the gravity increases , the electron pressure increases rapidly ( eq . [ pe ] ) , resulting in a higher h@xmath127 free - free absorption .
other effects of increasing the gravity are higher number densities supporting the molecular formation and the fact that molecules can be formed till higher temperatures , both increasing @xmath141 .
the final result is that the approximate absorption coefficient @xmath119 can either increase or decrease depending on the relative change of l@xmath136 and @xmath134 . using the approximate relations according to kjrgaard et al .
( 1982 ) ( see appendix b ) , the conclusion is reached that an increase of gravity leads to a decrease of the strength of the co , nh and cn lines and to an increase of the strength of the h@xmath117o lines , with no dependence for the oh and sio lines .
the increasing gravity does lead to a strengthening of the h@xmath117o lines and also of the oh lines while it diminishes the @xmath142 ratio of co , nh , sio and cn ( fig .
[ overall ] ) . from the approximate relations for @xmath119 no quantitative conclusion on the parameter dependence can thus be made .
an increase in mass yields the inverse ( but smaller ) effect on the partial pressure ( and so on the line absorption coefficient @xmath141 ) as an increase in gravity . for the computations with a higher metallicity , [ atom / h ]
has been increased from 0.00 dex to 0.15 dex for all atoms except for h , he , c , n and o. because an increase in metal abundance leads to a decrease in gas pressure at each optical depth ( as @xmath134 increases ) , the partial pressures of the molecules co , oh , n@xmath117 , h@xmath117o and cn consequently decrease , while p(sio ) and p(tio ) are somewhat higher because of the higher abundance of si and ti respectively .
an increase in the overall metal abundance increases the amount of electrons ( @xmath135 ) and as a consequence also the value of @xmath134 formed by h@xmath127 .
this increase of @xmath134 dominates the changes in the spectrum : all molecular features become fainter at increased metallicities ( with c , n and o kept constant ) ( fig .
[ overall ] ) .
this can also be seen in the equations for @xmath119 ( see appendix b ) for the extreme case in which all carbon is present in the form of co with t@xmath143 k. the strong metallicity dependence of h@xmath117o in the approximate relation for @xmath119 may be well visible , though this increase ( with decreasing z ) is small for weak h@xmath117o bands .
the impact of changing the microturbulent velocity is most important for the co lines with a large equivalent width .
when a line is saturated , increasing @xmath8 widens the wavelength range covered by the absorption and reduces the saturation , thus increasing the total absorption . in the line center of an unsaturated line , a smaller microturbulence corresponds to a higher absorption coefficient at that frequency ( see fig .
[ overall ] ) . decreasing the microturbulent velocity from 2 @xmath10 to 1 @xmath10diminishes the co first overtone absorption by maximum 10% and the co fundamental by up to 6% , the sio first overtone by up to 4% , and the sio fundamental by up to 3% , the fundamental band of oh by up to 3% and the ro - vibrational stretching modes ( @xmath144 and @xmath145 ) and bending mode ( @xmath146 ) of h@xmath117o by up to 0.5% .
red giant atmospheres display altered c , n , and o abundances and isotopic ratios ( in particular @xmath11 ) , due to dredge - up episodes .
suppose that the @xmath11 ratio is changed from 10 ( case i ) to 4.26 ( case ii ) . if a @xmath24co line is saturated in case i , its equivalent width will decrease very few from case i to case ii . if we assume the @xmath25co line to be very weak in case i , its equivalent width will increase very much from case i to case ii by a factor nearly equal to the new / old isotopic ratio .
therefore it is understandable that a decrease in @xmath11 will consequently cause @xmath119(co ) to increase , although this increase may be very small if the co bands are weak ( fig .
[ overall ] ) . when the ratio c / o increases ( but remains @xmath147 1 ) , the structure of the atmosphere is affected through changes in line blanketing . more co will be formed and less oxygen will be left ( after co formation ) to form oxygen - based molecules . hence , as illustrated in fig .
[ overall ] , the total absorption of co will increase , while it will decrease for sio , oh and h@xmath117o .
increasing the oxygen abundance gives roughly the reverse effect , but the total co absorption hardly changes .
the largest effect of an increasing nitrogen abundance is an increase of the nh and cn line strengths .
[ 0.5 ] fig . [ startvalues ] reveals several discrepancies . as outlined in sect .
1 , the origin of these discrepancies can be either a calibration problem or a problem related to the generation of the synthetic spectrum . precisely because this research involves both theoretical developments on the model spectra and calibration improvements of the spectral reduction method
, one has to be extremely careful not to confuse technical detector problems with astrophysical issues .
therefore , several points have to be taken into account when scrutinizing these discrepancies .
first of all , one has to take into account that the calibration process is a delicate one . due to the high flux level of the observations ,
dark - current subtraction does not influence the final spectrum much , but memory effects affect the reliability of band-2 observations .
therefore , mainly the band-1 data were used for the determination of the stellar parameters .
the final result is however very sensitive to the relative spectral response function ( rsrf ) used .
the rsrf was intensively measured in the lab before launch by use of a cryogenic blackbody source with temperatures between 30 and 300k .
after the launch of iso these intermediate rsrfs were corrected for broad - band residues detected by comparing reference spectral energy distributions ( seds ) synthetic spectra ( kurucz 1993 ; van der bliek et al .
1998 ) and composites of m. cohen ( cohen et al . 1992 , 1996a ; witteborn et al . 1999 ) with sws observations processed with these intermediate rsrfs using stellar candles of different ir spectral signature . by using a cubic spline smoothing algorithm and sigmaclipping
the risk was minimized of introducing false spectral features and one also avoids possible bias by spectral features of one type of source ( vandenbussche , 1999 ) .
the absolute flux level is determined by use of photometric data and composite seds of various sources .
the flux calibration accuracy is estimated to be 5 7% in band 1 , 7 12% in band 2a , 7 15% in band 2b and 11 25% in band 2c , the worst accuracy being located at the edges .
the spectral cover from a0
m7 was highly necessary , since the different structures of the various selected stars make it possible to decide upon the quality of the used rsrfs . to test our findings , high - resolution observations of two stars
were also studied in detail . for a discussion about the quality of the rsrfs
, we refer to the forthcoming papers which will describe the other 15 stars in our sample . in order to elucidate problems with the theoretical atmospheric structure ,
it is important to know the relative importance of the different molecules ( see fig .
[ atdiv ] ) and the influence of the stellar parameters on the total absorption by these molecules ( see sect . [ param ] ) .
[ atdiv ] shows clearly that especially the molecules co , oh and sio are helpful in determining the atmospheric parameters . in spite of the moderate resolution of iso - sws , we will demonstrate that one can pin down the stellar parameters of cool giants very accurately from these data .
this is due to the large wavelength range of iso - sws , where different important molecules determine the spectral signature .
since these different spectral features do each react in another way on a change of one of the several heterogeneous stellar parameters , it is possible to improve on the stellar parameters . using a metallicity of z @xmath148 yields , e.g. , a synthetic spectrum which has too low a flux density in the wavelength range from 2.6 2.8 @xmath19 m , with the line contrast @xmath142 being too strong .
[ overall ] turned out to be extremely useful to ascertain the gravity : the inverse influence of the gravity on the oh lines compared to the co and sio lines , together with the steep slope in band 1b and band 1d make it possible to pin down the logarithm of the gravity within 0.15 dex ! because the microturbulent velocity @xmath8 does not only act on the co lines , but also on the strongest sio and oh lines , one may disentangle a wrong microturbulence and a wrong carbon abundance . in order to achieve the highest possible agreement between the data and the different synthetic spectra , a statistical approach is proposed .
a choice was made for the kolmogorov - smirnov test .
this well - developed goodness - of - fit criterion is applicable for a broad range of comparisons between two samples , where the kind of differences which occur between the samples can be very diverse .
an elaborate discussion about the kolmogorov - smirnov statistics can be found in pratt & gibbons ( 1981 ) and hjek ( 1969 ) . in this goodness - of - fit test
a parameter , @xmath149 , is defined in the present case as the ratio of the iso - sws flux to the synthetic flux at frequency point @xmath150 .
these parameters @xmath149 are assumed to be independent random variables with @xmath151 and @xmath152 , with a constant unspecified variance @xmath153 .
the random variables @xmath154 ( with @xmath155 the total number of frequency points ) are then asymptotically behaving like @xmath156 and hence are asymptotically equidistributed on the interval @xmath157 $ ] . these variables @xmath158 will be formally compared with the quantiles of the uniform distribution @xmath159 , @xmath160 , using a modification of the kolmogorov - smirnov test .
taking the unspecified distribution function of the sample @xmath161 as @xmath162 , the null hypothesis is @xmath163 .
the kolmogorov - smirnov test is based on the maximum difference .
specifically , the test statistics are given by @xmath164 one rejects @xmath163 if @xmath165 is ` too large ' in comparison to an appropriate critical value .
practically , our modification involves the calculation of the value @xmath166 , defined as @xmath167 firstly , we will compute the asymptotical distribution function of @xmath166 . by defining @xmath168 one
has that @xmath169 and @xmath170 .
so , @xmath171 by defining @xmath172 $ ] , where @xmath173 $ ] is the greatest integer less than or equal to @xmath174 , eq .
( [ eqlong ] ) becomes @xmath175 } x_k - \frac{[nt]}{n } \left ( \frac{\sigma}{\sqrt{n } } \sum\limits_{k=1}^{n } x_k \right ) } { \displaystyle 1 + \frac{\sigma}{n } \sum\limits_{k=1}^{n } x_k}\right|.\ ] ] using the strong law of large numbers @xmath176 and donsker s theorem ( billingsley , 1968 , p. 137 ) @xmath177 } x_k \stackrel{\mathcal{d}}{\to } w(t),\ ] ] where @xmath178 , @xmath179 , is standard brownian motion , i.e. , @xmath178 is a gaussian process with @xmath180 , independent increments , and @xmath181 has a normal distribution with mean zero and variance @xmath182 , one sees that @xmath166 converges in distribution to @xmath183 here @xmath184 , @xmath179 , is known as a brownian bridge .
the definition of @xmath185 implies that @xmath185 is a gaussian process with @xmath186 and + @xmath187 if @xmath188 . as a consequence @xmath189 which implies that the asymptotic distribution function of @xmath190 is given by the kolmogorov - smirnov distribution @xmath191 ( billingsley , 1968 , p. 105 ) . for a test with significance level
@xmath0 one rejects the null hypothesis when @xmath192 , where @xmath193 for @xmath194 one has @xmath195 and for @xmath196 one has @xmath197 .
so , a good approximation to the level-@xmath0 kolmogorov - smirnov test is to reject a fit if the empirical distribution function exits from the bounds @xmath198 ( see , e.g. , fig .
10.2 in brockwell & davis ( 1991 ) )
. the lower the @xmath166-value , the better the accordance between the observed data and synthetic spectrum .
it is clear that @xmath199 gives us not only information about the correspondence between the observed and synthetic spectra , but also about @xmath200 .
since @xmath149 is the ratio of the iso - sws flux ( @xmath201 ) to the synthetic flux ( @xmath202 ) , it is obvious that @xmath203 and thus @xmath204 the unknown variance @xmath153 is thus an indication on the variance of the observed data . the kolmogorov - smirnov test _ globally _ checks the goodness of fit of the observed and synthetic spectra .
an advantage of this test is that one very discrepant frequency point ( e.g. due to a wrong oscillator strength ) only mildly influences the final result . to avoid too high an influence of the less reliable calibration at the band edges ,
the parameter @xmath166 was computed for each sub - band separately . in each sub - band ,
several sub - intervals were considered to the main absorber in the relevant wavelength range , and the parameter @xmath166 was computed for each of this sub - intervals .
the resulting ( @xmath205)-figures turned out to be very useful to reveal systematic problems ( see , e.g. , fig .
[ statistic ] ) .
[ 0.8 ] very low @xmath166-values ( @xmath206 ) were achieved , proving a good relation between observed and synthetic spectra .
the objective @xmath166-parameters , which are - in se - deviation estimating parameters , were then further used to improve the stellar parameters .
a sensible rule of thumb would then be to select the synthetic spectrum with the lowest @xmath166-values , among those that are consistent with the null hypothesis .
in addition , it is a useful sensitivity analysis to compare various spectra with acceptable @xmath166-values in key characteristics .
the high flux accuracy enables us to distinguish between @xmath207 and @xmath208 ( see fig .
[ statistic ] ) , although both values are in the same region of accepting the null hypothesis . from the sensitivity analysis
, it seemed to be useful to determine a maximum acceptable @xmath166-value ( i.e. a minimal deviation ) as an objective criterion .
a synthetic spectrum is accepted to represent the true stellar parameters of @xmath0 tau ( and the other stars ) when the @xmath166-values are lower than the values given in column 2 of table [ beta ] .
these empirical @xmath209-values give , as already mentioned ( eq .
( [ sigma2 ] ) ) , an indication on the uncertainty on the observational spectrum determined by e.g. the pointing accuracy , the pointing jitter of iso , uncertainties on the rsrfs , problematic dark - current subtraction , ... in fig .
[ stdevat ] the standard deviation , obtained when rebinning the oversampled spectrum of @xmath0 tau to the expected resolution , of each bin is shown .
obviously , the iso - sws spectrum is less accurate in band 1a than in the other bands .
this is reflected in the higher maximum acceptable @xmath166-value in table [ beta ] . in column 3 of table [ beta ] ,
the final @xmath166-values for @xmath0 tau are given . ' '' ''
sub - band/ & maximum acceptable & @xmath166-value + ' '' '' molecule & @xmath166-value & for @xmath0 tau + ' '' ''
1a & 0.06 & 0.040 + 1b & 0.05 & 0.034 + 1d & 0.04 & 0.036 + ' '' ''
1e & 0.04 & 0.027 + ' '' '' co ( 1a ) & 0.06 & 0.040 + co ( 1b ) & 0.05 & 0.039 + oh ( 1b ) & 0.04 & 0.024 + oh ( 1d ) & 0.04 & 0.036 + oh ( 1e ) & 0.03 & 0.025 + ' '' '' sio ( 1e ) & 0.03 & 0.022 + [ 0.3 ] due to the smaller weight which are given automatically to small features , the traditional comparison between observed and synthetic spectra by eye - ball fitting is still necessary as a complement to this kolmogorov - smirnov method in order to reveal systematic errors in those features . the final error bars on the atmospheric parameters
are then estimated from 1 . the intrinsic uncertainty on the synthetic spectrum ( i.e.the possibility to distinguish different synthetic spectra at a specific resolution i.e. there should be a relevant difference in @xmath166-values ) which is thus dependent on both the resolving power of the observation and the specific values of the fundamental parameters , 2 . the uncertainty on the iso - sws spectrum which is directly related to the quality of the iso - sws observation , 3 .
the value of the @xmath166-parameters in the kolmogorov - smirnov test and 4 .
the still remaining discrepancies between observed and synthetic spectrum .
however , no exact formula can be given to compute the error bars due to the dependence of the several parameters .
when a quantity @xmath210 depends on other independent parameters , e.g. @xmath211 and @xmath212 , the uncertainty @xmath213 is estimated from the expressions @xmath214 or , equivalently @xmath215 the power of the kolmogorov - smirnov test has already been proven in the field of astronomy , even multidimensional versions of the kolmogorov - smirnov test are developed for astronomical purposes ( fasano & franceschini 1987 ; gosset 1987 ) .
fasano & franceschini ( 1987 ) demonstrate even that the kolmogorov - smirnov test is a much more goodness - of - fit test than the often used @xmath216-test .
it is , however , the first time that this test is used to evaluate synthetic spectra with respect to observational data . to summarize the strategy : + 1 ) from the large set of stars and two high - resolution observations ,
it is possible to indicate calibration problems ( see forthcoming paper ) ; + 2 ) in the comparison between observed and synthetic spectrum , some discrepancies were clearly visible .
the knowledge on the relative importance of the different molecules ( fig .
[ atdiv ] ) and on the influence of the various stellar parameters ( fig .
[ overall ] ) makes it possible to elicit the origin of these discrepancies .
this is due to the presence of different molecules in this large wavelength range , whose spectral shape and strength are a very good diagnostic tool to reveal the stellar parameters ; + 3 ) once a high level of accuracy is achieved , the kolmogorov - smirnov test can be used for further refinement . from the ( @xmath205 ) plots ,
one can deduce systematic errors , yielding an indication on inaccurate stellar parameters . by using the results illustrated in fig .
[ overall ] , one can improve the stellar parameters .
the stellar parameters , used to generate a synthetic spectrum , are taken to represent the ` true ' stellar parameters , once the @xmath166-values are lower than a certain maximum value .
our best fit involves a synthetic spectrum with stellar parameters t@xmath1 = 3850 k , @xmath3 g = 1.50 , m = 2.3 m@xmath6 , z = @xmath120 , @xmath8 = 1.7 @xmath10 , @xmath11 = 10 , @xmath13(c ) = 8.35 , @xmath13(n ) = 8.35 , @xmath13(o ) = 8.83 and @xmath17 = 20.77 mas ( see fig . [ result ] ) .
the uncertainties are estimated to be @xmath121t@xmath1 @xmath217 k , @xmath122 g@xmath218 , @xmath121@xmath8
@xmath219 @xmath10 , @xmath121z @xmath220 , @xmath123(c ) @xmath220 , @xmath123(n ) @xmath221 , @xmath123(o ) @xmath222 , @xmath121@xmath11 @xmath223 and @xmath224 . the quoted accuracy of 0.03 mas does not take the absolute photometric flux accuracy into account , which is 5% in band 1 , resulting in @xmath225 mas . [
0.65 ] due to the minor influence of the stellar mass , the uncertainty on this parameter , which follows from the construction of a synthetic spectrum , is quite large .
but by using the parallax this uncertainty is constrained to @xmath121 m @xmath226 m@xmath6 .
the parallax measurement by hipparcos for @xmath0 tau is @xmath227 mas , which corresponds to a distance of @xmath228 pc . with an angular diameter @xmath17 of @xmath229 mas ,
a radius r @xmath230 r@xmath6 is found , which combined with @xmath3 g @xmath4 , implies a mass of @xmath231 m@xmath6 .
a refinement in the mass determination was proposed by el eid ( 1994 ) , who noted a correlation between the @xmath232o/@xmath233o ratio and stellar mass for evolved stars . using this ratio in conjunction with evolutionary calculations ,
el eid ( 1994 ) found a mass of 1.5 m@xmath6 for @xmath0 tau .
a comparison of the deduced luminosity and effective temperature with evolutionary tracks ( cfr .
mcwilliam 1990 ) indicates a mass of 2.0 m@xmath6 for @xmath0 tau .
the accuracy of the method can be tested by making a grid around the resultant final stellar parameters .
this is illustrated for @xmath0 tau in table [ grid ] .
the @xmath166-values , corresponding to the final stellar paramters , for the different sub - bands are @xmath234 , @xmath235 , @xmath236 , @xmath237 ( see table [ beta ] ) . in table
[ grid ] , the @xmath166-values are given for a change of one of the main stellar parameters , being the temperature , the gravity , the metallicity , the carbon abundance , the nitrogen abundance and the oxygen abundance . from this table , one can deduce that there are four other acceptable sets of parameters , indicated by a grey box . the final stellar parameters as mentioned above were chosen to represent the ` true ' atmospheric parameters of @xmath0 tau , due to the lower @xmath166-value in band 1a , where the important molecule co is absorbing .
r||cccc ' '' '' & @xmath238 & @xmath239 & @xmath240 & @xmath241 + @xmath242 & 0.060 & 0.050 & 0.040 & 0.040 + ' '' '' @xmath243 & 0.040 & 0.034 & 0.036 & 0.027 + ' '' ''
t@xmath1 = 3710 & 0.080 & 0.050 & 0.059 & 0.027 + & & & & + 3920 & 0.031 & 0.043 & 0.084 & 0.028 + ' '' ''
3990 & 0.044 & 0.046 & 0.104 &
0.041 + ' '' '' @xmath3 g = 1.20 & 0.037 & 0.093 & 0.071 & 0.015 + 1.35 & 0.034 & 0.063 & 0.064 & 0.018 + & & & & + ' '' '' 1.80 & 0.051 & 0.055 & 0.032 & 0.027 + ' '' '' z = @xmath244 & 0.037 & 0.094 & 0.075 & 0.013 + @xmath113 & 0.034 & 0.062 & 0.066 & 0.015 + & & & & + ' '' '' 0.15 & 0.050 & 0.054 & 0.029 & 0.030 + ' '' ''
@xmath13(c ) = 8.05 & 0.047 & 0.132 & 0.040 & 0.018 + 8.20 & 0.042 & 0.065 & 0.045 & 0.021 + 8.50 & 0.036 & 0.093 & 0.063 & 0.031 + ' '' '' 8.65 & 0.065 & 0.174 & 0.085 & 0.044 + ' '' ''
@xmath13(n ) = 7.85 & 0.043 & 0.032 & 0.051 & 0.019 + 8.10 & 0.041 & 0.031 & 0.051 & 0.021 + 8.60 & 0.034 & 0.032 & 0.055 & 0.025 + ' '' '' 8.85 & 0.031 & 0.034 & 0.059 & 0.028 + ' '' ''
@xmath13(o ) = 8.63 & 0.057 & 0.049 & 0.096 & 0.046 + 8.73 & 0.035 & 0.040 & 0.075 & 0.034 + & & & & + ' '' '' 8.03 & 0.070 & 0.043 & 0.028 & 0.019 + according to fig . 20 in bessell et al .
( 1998 ) , @xmath0 tau , with a ( @xmath245 ) colour index of 3.67 , has a bolometric correction bc@xmath246 of @xmath247 . from the bolometric magnitude @xmath248 , the distance @xmath249 , the assumption that the interstellar extinction may be neglected , and by adapting the absolute bolometric magnitude of the sun @xmath250 ( according to bessell et al .
1998 ) , one deduces the luminosity of @xmath0 tau l @xmath251 l@xmath6 .
a radius r of @xmath252 r@xmath6(calculated from the distance @xmath253 and the angular diameter @xmath17 ) yields an effective temperature of @xmath254 k. this is in very good agreement with the effective temperature deduced from the iso - sws spectrum , being t@xmath1 @xmath2 k. this latter t@xmath1 , in conjunction with l @xmath255l@xmath6 , would give r @xmath256
r@xmath6and @xmath17 @xmath257 mas .
it is clear that a gravity of @xmath3 g = @xmath258 dex agrees better with a ` physical ' gravity ( @xmath3 g@xmath259 from m = 2 m@xmath6 and r = 44.6 r@xmath6 ) determined by the mass m , the radius r or the effective temperature t@xmath1and the luminosity l and/or evolutionary tracks than with a ` spectroscopic ' gravity of e.g. lambert & ries ( 1981 ) , kovcs ( 1983 ) and luck & challener ( 1995 ) ( see appendix a ) .
it is well known that the accuracy of spectroscopic determinations of gravities from ionization equilibria or molecular equilibria for individual stars is not very good for red giants ( cf .
, e.g. , trimble & bell 1981 , brown et al .
1983 , smith & lambert 1985 ) .
smith & lambert ( 1985 ) suggested that the low spectroscopic gravity @xmath3 g = 0.8 may be due to non - lte effects on iron lines .
after corrections for these effects , smith & lambert ( 1990 ) finally determined values for the various atmospheric parameters , with which our results agree very well .
our method for gravity determination from infrared spectra appears to be much less affected by nlte effects , being based on a broad range of diagnostics over a large spectral interval .
a great advantage of the used kolmogorov - smirnov test , is also that several deviation estimating parameters @xmath166 are calculated , which are not independent and are each of them related to the specific pattern of behaviour of absorption of the different molecules .
e.g.the abundance of carbon is mainly determined from the band-1a and band-1b data , while a wrong gravity is traced from the @xmath260-values .
simulations with other ( @xmath3 g , @xmath13(c))-combinations never reached the same level of agreement as for the ( @xmath3 g = 1.50 , @xmath13(c ) = 8.35)-combination . using e.g. @xmath3 g
= 1.35 requires a lower carbon abundance ( than @xmath13(c ) = 8.35 ) in order to fit the co line strengths .
the best fit then is obtained for @xmath13(c ) = 8.25 , with @xmath238 = 0.033 and @xmath239 = 0.024 .
the corresponding @xmath240-value is then , however , 0.056 arising from too steep a continuum .
the mass m , luminosity l and radius r are the three fundamental parameters of a ( spherical ) star where two of these may be replaced by the surface gravity g = @xmath261m / r@xmath262 and the effective temperature @xmath200t@xmath1@xmath263=l/(4@xmath264r@xmath262 ) .
both the mass and the luminosity are measurable physical quantities , whereas the radius of a star is a fictitious quantity because a star is a gaseous sphere and does not have a sharp edge .
the relevant observable quantity is the center - to - limb variation ( = clv ) of intensity or limb darkening .
the sun is the only star whose clv one can observe directly .
direct diameter determinations are based on a standard procedure which evaluates interferometric , lunar occultation or binary eclipse data by determining a radius position on the base of a parameterized approximation ( e.g.uniform disk ) of a model - predicted limb - darkening curve ( see e.g.scholz 1998 ) .
the radius is then taken at @xmath265 ( @xmath266 , with @xmath267 the angle between the radius vector and the line of sight ) .
published limb - darkening coefficients ( for an overview , see scholz ( 1998 ) ) are then needed to convert to the limb - darkened corrected angular diameter . in table [ literature ] , this limb - darkened angular diameter is listed for the direct angular diameter determinations , except when otherwise mentioned . the angular diameter value derived by us from the iso - sws data is however a ` theoretical ' radius deduced from an indirect method .
it corresponds to the definition @xmath200t@xmath1@xmath263 = l/(4@xmath264r(@xmath268)@xmath262 ) and @xmath3 g = gm / r(@xmath268)@xmath262 ( plez et al .
baschek et al .
( 1991 ) have summarized various radius definitions .
one of the most common theoretical definitions for the radius is the _ optical depth radius_. assuming that most of the flux emerges from a layer with optical depth @xmath269 , is the base for defining a radius r@xmath270 = r(@xmath271 ) , which depends of course on the extinction coefficient at this wavelength . in fig . 2 of scholz ( 1998 )
one can see clearly that there is no trivial correlation between the clv shape and the position of this @xmath271 radius .
the angular diameter ( and radius ) obtained from the iso - sws data should therefore be only compared to other indirect spectrophotometric angular diameter values , resulting from e.g.comparing the observed flux at a distance d from the star with the surface flux , the surface brightness method ( introduced by barnes & evans 1976 ) or the infrared or rayleigh - jeans flux method ( from blackwell & shallis 1977 ) .
the quadratic dependence of the observed flux on the angular diameter makes an accurate determination of @xmath17 from the iso - sws data possible . with the exception of a few references given in the catalogue of fracassini et al .
( 1988 ) , the angular diameter @xmath17 @xmath272 mas is in very good agreement with the other indirect determined angular diameter values .
the carbon abundance is set by the strength of the co absorption , while the strength of the oh lines yields the oxygen abundance .
due to the weakness of the nh and cn lines in @xmath0 tau ( see fig .
[ atdiv ] ) , it was impossible to determine @xmath13(n ) directly from the sws spectrum . since the chemical equilibrium detemines however the shape of the synthetic spectrum , a too inaccurate nitrogen abundance would result in high @xmath166-values , which did not happen when using @xmath13(n ) = 8.35 ( see table [ grid ] ) .
the large dependence of the co line - width on @xmath13(c ) implies an uncertainty @xmath273 dex .
this can also be seen in table [ grid ] : changing the carbon abundance by 0.15 dex results in e.g.too high a @xmath239-value so that only for @xmath274 no other acceptable set of @xmath166-values appears in table [ grid ] .
due to , however , the larger standard deviation after rebinning in band 1a and the beginning of band 1b than in the other sub - bands , the total uncertainty on @xmath274 is taken to be 0.20 dex . from table [ grid ] , one also can deduce that the error bars on @xmath13(n ) and @xmath13(o ) , being 0.25 dex and 0.15 dex respectively , are realistic values .
the carbon , nitrogen and oxygen abundance , together with the low isotopic @xmath11 ratio of @xmath0 tau are an argument for @xmath0 tau being on the first giant branch . with a luminosity of 382 l@xmath6 and t@xmath1 @xmath275 k ,
@xmath0 tau lies slightly above the position where core helium - burning starts for stars with 1.0 m@xmath6 @xmath276 m @xmath276 2.0 m@xmath6 and has an age of @xmath277 5 gyr in the evolutionary diagram of claret & gimenez ( 1995 ) .
consequently there are two possibilities for the evolutionary stage of @xmath0 tau : + ( i ) @xmath0 tau is still on the first giant branch , or + ( ii ) @xmath0 tau has ignited helium - burning and either had not enough time to move into the region of the clump stars or is already on the asymptotic giant branch .
+ following the discussion of kovcs ( 1983 ) , it seems impossible to decide whether @xmath0 tau is in phase ( i ) or in phase ( ii ) .
so far , the low @xmath11 ( @xmath116 ) value can not be explained by standard evolutionary models for low - mass stars .
charbonnel ( 1994 ) introduced an extra - mixing process that occurs later on the giant branch and produces an additional decrease of the surface @xmath11 and c / n values .
this extra - mixing process should be only efficient when the hydrogen - burning shell has reached the discontinuity in molecular weight .
charbonnel ( 1994 ) held the process of zahn ( 1992 ) who proposed a consistent picture of the interaction between meridional circulation and turbulence induced by rotation in stars responsible for the required extra - mixing on the red giant branch of low mass stars .
we want to stress that the current stellar parameters are obtained by using the 1-dimensional models as described in sect .
[ models ] .
since there are indications of problems with the temperature distribution in the outermost layers of the theoretical models ( see forthcoming paper ) , a future project is to extend the theoretical models to include the ability to conduct temperature perturbations in order to simulate convection , the presence of a chromosphere or a dust shell and a change in opacity . using these improved models
, it will be interesting to study the changes in the abundance ( especially carbon , nitrogen and oxygen ) needed to match the line strengths of co ( fundamental and first overtone)lines , sio ( fundamental and first overtone ) lines , h@xmath117o fundamental lines and oh fundamental lines in the iso - sws spectra and high - resolution fourier transform spectrometer ( fts ) spectra .
in addition , by studying the effect of changing the microturbulence @xmath8 , it might be possible to identify a typical region associated with a specific spectral type in the 3-dimensional ( @xmath278 , @xmath8 , @xmath13)-plane , with @xmath13 being the abundance of carbon , nitrogen of oxygen .
so far , it was still not possible to identify such a region from the many different values for the stellar parameters of red giants derived from different methods and/or data ( cf . ,
e.g. , lambert & ries ( 1981 ) versus mcwilliam ( 1990 ) ) .
only a systematic study may reveal such a region .
the composite sed of cohen et al .
( 1992 ) is also plotted on fig .
[ result ] and lies @xmath279% higher than the sws data , which is well within the quoted flux calibration accuracy .
cohen et al .
( 1992 ) noted however a global accuracy of @xmath280% for this @xmath0 tau sed and a local accuracy which is somewhat lower .
the accuracy quoted by cohen et al .
( 1992 ) thus appears slightly optimistic .
the remaining discrepancies seen in band 2 are due to the memory effects which affect the reliability of observations in this band .
the obtained synthetic spectrum will be used as a test vehicle for the memory correction procedures which are currently under development .
the full scan aot01 speed-4 iso - sws observation of @xmath0 tau has been used to determine the stellar parameters of this k5 giant . due to the complementarity of the way parameters affect the spectrum in the wavelength range from 2.38 12 @xmath19 m like the presence of different molecules which are each dependent in another way on the several stellar parameters it was possible to pin down the atmospheric parameters with a high accuracy .
this resulted in the following parameters t@xmath1 = @xmath281 k , @xmath3 g = @xmath258 , m = @xmath231 m@xmath6 , z = @xmath282 , @xmath8 = @xmath283 @xmath10 , @xmath11 = @xmath284 , @xmath13(c ) = @xmath285 , @xmath13(n ) = @xmath286 , @xmath13(o ) = @xmath287 and @xmath17 = @xmath288 mas .
these parameters were compared with and discussed w.r.t .
the results listed by other authors , one of the most striking result being that one should rely more upon a physically determined gravity than upon a spectroscopic gravity .
the good consistency between the stellar parameters deduced from the iso - sws data , colours , high - resolution ( optical ) spectra , irfm , ... proves that it is possible to extract good quantative informations from these intermediate - resolution iso - sws spectra !
\1 . mc william ( 1990 ) based his results on high - resolution spectroscopic observations with resolving power 40000 . + the effective temperature was determined by empirical and semi - empirical results found in the literature and from broad - band johnson colours .
the gravity was ascertained by using the well - known relation between g , t@xmath1 , the mass m and the luminosity l , where the mass was determined by locating the stars on theoretical evolutionary tracks .
so , the computed gravity is fairly insensitive to errors in the adopted l. high - excitation iron lines were used for the metallicity [ fe / h ] in order that the results are less spoiled by non - lte effects .
the author refrained from determining the gravity in a spectroscopic way ( i.e. by requiring that the abundance of neutral and ionized species yield the same abundance ) because _ ` a gravity adopted by demanding that neutral and ionized lines give the same abundance , is known to yield temperatures which are @xmath289 k higher than found by other methods .
this difference is thought to be due to non - lte effects in fe i lines . ' _ by requiring that the derived iron abundances , relative to the standard 72 cyg , were independent of equivalent width , the microturbulent velocity @xmath8 was found .
lambert & ries ( 1981 ) have used high - resolution low - noise spectra .
the parameters were ascertained by demanding that the spectroscopic requirements ( ionization balance , independence of the abundance of an ion versus the excitation potential and equivalent width ) should be fulfilled .
the effective temperature was found from the fe i excitation temperature and the model atmosphere calibration of the excitation temperature as a function of t@xmath1 . as quoted by ries ( 1981 ) , harris et al .
( 1988 ) and luck & challener ( 1995 ) their t@xmath1 and @xmath3 g are too high and should be lowered by 240 k and 0.40 dex , respectively .
the isotopic ratio @xmath11 was taken from tomkin et al .
( 1975 ) , while the luminosity was estimated from the k - line visual magnitude @xmath290 given by wilson ( 1976 ) and the bolometric correction bc by gustafsson & bell ( 1979 ) .
the abundances of carbon , nitrogen and oxygen were based on c@xmath117 , [ o i ] and the red system cn lines respectively .
luck & challener ( 1995 ) wondered whether the nitrogen abundance [ n / fe ] = @xmath291 quoted reflects a typographic error and should rather be [ n / fe]=+0.20 , resulting in @xmath13(n)=7.86 , which is more in agreement with being a red giant branch star .
the effective temperature t@xmath1 and the angular diameter @xmath17 given by blackwell et al .
( 1990 ) were determined by the infrared flux method ( irfm ) , a semi - empirical method which relies upon a theoretical calibration of infrared bolometric corrections with effective temperature .
one expects that the irfm should yield results better than 1% for the effective temperature and 2 3% for the angular diameter .
the final effective temperature is a weighted mean of t(j@xmath292 ) , t(k@xmath292 ) and t(l@xmath292 ) , with j@xmath292 at 1.2467 @xmath19 m , k@xmath292 at 2.2135 @xmath19 m and l@xmath292 at 3.7825 @xmath19 m .
the stellar parameters quoted by tsuji ( 1986 , 1991 ) were based on the results of tsuji ( 1981 ) , in which the temperature was determined by the irfm method . a mass of 3 m@xmath6 was assumed to ascertain the gravity .
tsuji ( 1986 ) has used high - resolution fts spectra of the co ( first overtone ) lines to determine @xmath13(c ) and @xmath8 by assuming that the abundance should be independent of the equivalent width of the lines . in tsuji ( 1991 )
co lines of the second overtone were used , where a standard analysis yielded results of @xmath13(c ) = 8.39 and a linear analysis of weak lines led to @xmath13(c ) = 8.31 .
high - resolution spectra of oh ( @xmath293 ) , co(@xmath294 ) , co(@xmath293 ) and cn(@xmath295 ) were obtained by smith & lambert ( 1985 ) .
they have used ( @xmath245 ) colours and the calibration provided by ridgway et al .
( 1980 ) to determine t@xmath1 . using the spectroscopic requirement that @xmath13(fe i ) = @xmath13(fe
ii ) yields @xmath3 g = 0.8 dex , which is too low for a k5iii giant .
they suggested that the reason for this low value is the overionization of iron relative to the lte situation .
they then have computed the surface gravity by using a mass estimated from evolutionary tracks in the h - r diagram .
the metallicity was taken from kovcs ( 1983 ) and for the microturbulence they used fe i , ni i and ti i lines , demanding that the abundances are independent of the equivalent width . using the molecular lines , they determined @xmath13(c ) , @xmath13(n ) , @xmath13(o ) and @xmath11 .
harris & lambert ( 1984 ) have taken t@xmath1 , @xmath3 g and @xmath8determined by dominy , hinkle & lambert in 1984 , a reference which we could not trace back .
the isotopic ratio @xmath11 was adopted from tomkin et al .
the carbon abundance was found by fitting weak @xmath24c@xmath232o lines at 1.6 , 2.3 and 5 @xmath19 m .
\8 . kovcs ( 1983 ) obtained observations at the 1.52 m telescope of eso at la silla . by using different ir colour indices
( @xmath296 , @xmath297 , @xmath298 , @xmath245 , @xmath299 , @xmath300 for the broad - band photometry and @xmath301 , @xmath302 , @xmath303 , @xmath304 , @xmath305 for the narrow - band photometry ) the effective temperature was found .
the gravity and microturbulent velocity were determined in a spectroscopic way , where the abundance derived from strong and medium lines should equal the abundance derived from weak lines for the right value of @xmath8 .
the @xmath11 ratio was taken from tomkin et al .
( 1976 ) . using
a parallax of @xmath306 and @xmath17 of 24 mas , resulted in a radius of @xmath57 r@xmath6 , which corresponds to a luminosity l of @xmath307 l@xmath6 , while @xmath308 and a bolometric correction taken from johnson ( 1966 ) yields l = 413 l@xmath6 .
lambert et al . ( 1980 ) took model parameters from published papers which constrained the @xmath11 ratio .
the parameters for @xmath0 tau were based on tomkin et al .
( 1976 ) and lambert ( 1976 ) .
tomkin et al .
( 1976 ) have used ir colours for the determination of t@xmath1 and the microturbulence was ascertained by fitting the theoretical curve of growth to the @xmath24cn curve of growth ( @xmath24cn(2 - 0 ) around 8000 @xmath309 , weak @xmath24cn(4 - 0 ) lines around 6300@xmath309 and weak @xmath24cn(4 - 2 ) lines around 8430 @xmath309 ) . together with @xmath25cn lines around 8000@xmath309 , the isotopic ratio @xmath11 computed .
the gravities were estimated from the effective temperature , the mass and the luminosity . \10 . by fitting the engelke function ( engelke 1992 ) to the template of @xmath0 tau , cohen et al .
( 1996b ) could determine t@xmath1 and @xmath17 .
a gravity of 2.0 was adopted , though in cohen et al .
( 1992 ) a value of 1.5 taken from smith & lambert ( 1985) was used .
\11 . by taking the mean value of the different temperatures derived by calibrations with photometric indices ( @xmath310 , @xmath311 , @xmath296 , @xmath297 , @xmath298 , @xmath245 , @xmath299 ) fernndez - villacaas et al .
( 1990 ) have fixed the effective temperature .
for the gravity , the ddo photometry indices @xmath312 and @xmath313 were used .
the fe i lines served for the determination of @xmath8 .
van paradijs & meurs ( 1974 ) have adopted the angular diameter from currie et al .
( 1974 ) . to determine the effective temperature they used several continuum data , the curve of growth with the line strengths of fe i lines and the surface brightness .
requiring that the neutral and ionized lines of fe , cr , v , ti and sc gave the same abundance , yielded the gravity . gravity , parallax and angular diameter yielded the mass , while the luminosity was determined from the effective temperature , the parallax and the angular diameter .
van paradijs & meurs ( 1974 ) quoted that wilson ( 1972 ) found [ fe / h ] = @xmath314 .
tomkin & lambert ( 1974 ) obtained photoelectric scans of the red cn lines with a resolution of 0.05 @xmath309 for the red scans and 0.09 @xmath309 for the infrared scans .
the effective temperature and gravity were adopted from conti et al .
( 1967 ) , who have ascertained the effective temperature from the absolute magnitude derived from the k emission - line width and the gravity from the effective temperature , the mass and the luminosity .
the microturbulent velocity was determined using a curve of growth technique and the @xmath11 ratio was obtained directly from the horizontal shifts between the curves of growth for @xmath24cn and @xmath25cn lines .
also luck & challener ( 1995 ) have determined the effective temperature using photometric data ( ddo @xmath313 and @xmath312 , geneva @xmath315 , johnson @xmath245 , @xmath316 , @xmath296 and @xmath311 ) .
two methods were used to fix the surface gravity .
the first one determined the ` physical gravity ' by using the mass m and the radius r , with the mass m determined by t@xmath1 , l(@xmath290 , bc ) and theoretical evolutionary tracks and the radius r by t@xmath1 and l(@xmath290 , bc ) .
the ` spectroscopic gravity ' is based on the ionization balance of fe i and fe ii lines .
the difference between these two gravities was very large , being 1.0 dex !
when the fe ii oscillator strengths were modified to reflect a solar fe abundance of 7.50 ( instead of the used 7.67 ) , then the spectroscopic gravity scale would rise by + 0.25 dex ( still far less than 1.0 dex ) resulting in an increase of [ fe / h ] of @xmath317 dex .
they have quoted both pluses and minuses for the spectroscopic gravity and a single plus for the physical gravity .
the microturbulence was ascertained by forcing no dependence of abundance ( derived from individual fe i lines ) upon the equivalent width . in determining the carbon abundance using the c@xmath117 swan system lines and [ c i ] lines , the nitrogen abundance and isotopic ratio @xmath11 using cn lines and the oxygen abundance from [ o i ] lines
, they always found a better trend with temperature when the spectroscopic gravity was used and also a better agreement between the two carbon indicators .
when comparing their results with the ones of lambert & ries ( 1981 ) , they found that @xmath0 tau was always the most discrepant star .
their very low spectroscopic gravity , compared with the results of other authors using different methods and taking into account possible non - lte effects , may be an indication that a value of @xmath3 g @xmath21 1.5 is in better agreement with the real gravity of @xmath0 tau .
aoki & tsuji ( 1997 ) took the same stellar parameters as tsuji ( 1986 , 1991 ) , but they now used cn - lines to determine @xmath13(n ) and @xmath8 .
bonnell & bell ( 1993 ) constructed a grid based on the effective temperature t@xmath1 of manduca et al .
they used ground - based high - resolution fts spectra of oh and [ o i ] lines .
the requirement that the oxygen abundances determined from the [ o i ] and oh line widths agree amounts to finding the intersection of the loci of points defined by the measured widths in the ( [ o / h ] , @xmath3 g ) plane . for @xmath0
tau they found large discrepancies in the [ o i ] oxygen abundance , which leads to a spread of @xmath318 dex in @xmath3 g. the determination of @xmath8 was based on the oh - lines , but was hampered by a lack of weak lines from this radical . using a least square fit
, they found @xmath319 @xmath10 for the oh(@xmath320 ) lines and @xmath321 @xmath10 for the oh(@xmath322 ) lines , reflecting that the oh(@xmath320 ) sequence lines are formed at greater average depth relative to the oh(@xmath322 ) lines .
for the [ o i ] lines a microturbulent velocity of 1.5 @xmath10 or 2.0 @xmath10 was used .
ridgway et al .
( 1982 ) determined the limb - darkening - corrected angular diameter @xmath323 using the lunar occultation technique .
di benedetto & rabbia(1987 ) used michelson interferometry by the two - telescope baseline located at cerga . combining this angular diameter with the bolometric flux @xmath324 ( resulting from a directed integration using the trapezoidal rule over the flux distribution curves , after taking interstellar absorption into account ) they + found an effective temperature of 3970 k , which is in good agreement with the results obtained from the lunar occultation technique .
di benedetto ( 1998 ) calibrated the surface brightness - colour correlation using a set of high - precision angular diameters measured by modern interferometric techniques .
the stellar sizes predicted by this correlation were then combined with bolometric flux measurements , in order to determine one - dimensional ( t , v - k ) temperature scales of dwarfs and giants .
mozurkewich et al .
( 1991 ) used the markiii optical interferometer .
the uniform - disk angular diameter @xmath325 had a residual of 1% for the 800 nm observations and less than 3% for the 450 nm observations .
the limb - darkened diameter was then obtained by multiplying the uniform - disk angular diameter with a correction factor ( using the quadratic limb - darkening coefficient from manduca 1979 ) .
quirrenbach et al .
( 1993 ) have determined the uniform - disk angular diameter in the strong tio band at 712 nm and in a continuum band at 754 nm with the markiii stellar interferometer on mount wilson . because limb darkening is expected to be substantially larger in the visible than in the infrared
, the measured uniform - disk diameters should be larger in the visible than in the infrared .
this seems , however , not always to be the case . using the same factor as mozurkewich et al .
( 1991 ) we have converted their continuum uniform - disk value ( 19.80 mas ) into a limb - darkened angular diameter , yielding a value of 22.73 mas , with a systematic uncertainty in the limb - darkened angular diameter of the order of 1% in addition to the measurement uncertainty of the uniform - disk angular diameter ( davis 1997 ) .
volk & cohen ( 1989 ) mentioned a distance of @xmath326 pc .
the effective temperature was directly determined from the literature values of angular diameter measurements and total flux observations ( also from literature ) .
the distance was taken from the catalogue of nearby stars ( gliese , 1969 ) or from the bright star catalogue ( hoffleit 1982 ) .
blackwell et al .
( 1991 ) is a revision of blackwell et al .
( 1990 ) where the h@xmath127 opacity has been improved .
they investigated the effect of the improved h@xmath127 opacity on the irfm temperature scale and derived angular diameters . also here , the mean temperature is a weighted mean of the temperatures for @xmath327 , @xmath328 and @xmath329 .
relative to blackwell et al . (
1990 ) there was a change of temperature up to 1.4% and an decrease by 3.5% in @xmath17 .
\23 . by using a t@xmath1-(@xmath297 ) transformation of johnson ( 1966 ) , linsky & ayres ( 1978 ) have determined the effective temperature .
bell & gustafsson ( 1989 ) first determined the temperature from the johnson k band at 2.2 @xmath19 m by use of the irfm method . by comparison with temperatures deduced from the colours glass @xmath330 , @xmath331 , @xmath332 and @xmath333 ; cohen , frogel and persson @xmath330 , @xmath331 , @xmath332 and @xmath333 ; johnson @xmath298 , @xmath245 , @xmath299 and @xmath333 ; cousins @xmath296 , @xmath300 ; johnson and mitchell 13-colour and wing s near - infrared eight - colour photometry , they found that t@xmath1(irfm ) was @xmath334 k too high , by which they corrected the temperature .
the gravity and the metallicity were adopted from kovcs ( 1983 ) .
taylor ( 1999 ) prepared a catalogue of temperatures and [ fe / h ] averages for evolved g and k giants .
this catalogue is available at cds via anonymous ftp to cdsarc.u-strasbg.fr \26 .
burnashev ( 1983 ) has determined t@xmath1 , @xmath3 g and [ fe / h ] from narrow - band photometric colours in the visible part of the electromagnetic spectrum .
fracassini et al .
( 1988 ) have made a catalogue of stellar apparent diameters and/or absolute radii , listing 12055 diameters for 7255 stars .
only the most extreme values are listed .
references and remarks to the different values of the angular diameter and radius may be found in this catalogue . also here
these angular diameter values are given in italic mode when determined from direct methods and in normal mode for indirect ( spectrophotometric ) determinations .
perrin et al . (
1998 ) have derived the effective temperatures for nine giant stars from diameter determinations at 2.2 @xmath19 m with the fluor beam combiner on the iota interferometer .
this yielded the uniform - disk angular diameter of @xmath0 boo and @xmath0 tau of our sample .
the averaging effect of a uniform model leads to an underestimation of the diameter of the star .
therefore , they have fitted their data with limb - darkened disk models published in the literature .
the average result is a ratio between the uniform and the limb - darkened disk diameters of 1.035 with a dispersion of 0.01 .
this ratio could then also be used for the uniform - disk angular diameters of @xmath335 dra and @xmath166 peg , listed by di benedetto & rabbia ( 1987 ) , and @xmath0 cet , which was based on a photometric estimate .
several photometric sources were used to determine the bolometric flux , which then , in conjunction with the limb - darkened diameter , yielded the effective temperature .
engelke ( 1992 ) has derived a two - parameter analytical expression approximating the long - wavelength ( 2 60 @xmath19 m ) infrared continuum of stellar calibration standards .
this generalized result is written in the form of a planck function with a brightness temperature that is a function of both observing wavelength and effective temperature .
this function is then fitted to the best empirical flux data available , providing thus the effective temperature and the angular diameter .
blackwell & shallis ( 1977 ) have described the infrared flux method ( irfm ) to determine the stellar angular diameters and effective temperatures from absolute infrared photometry . for 28 stars ( including @xmath0 car , @xmath0 boo , @xmath0 cma , @xmath0 lyr , @xmath166 peg , @xmath0 cen a , @xmath0 tau and @xmath335 dra )
the angular diameters are deduced . only for the first four stars the corresponding effective temperatures are computed .
scargle & strecker ( 1979 ) have compared the observed infrared flux curves of cool stars with theoretical predictions in order to assess the model atmospheres and to derive useful stellar parameters .
this comparison yielded the effective temperature ( determined from flux curve shape alone ) and the angular diameter ( determined from the magnitudes of the fluxes ) .
the overall uncertainty in t@xmath1 is probably about 150 k , which translates into about a 9% error in the angular diameter .
manduca et al .
( 1981 ) have compared absolute flux measurements in the 2.5 5.5 @xmath19 m region with fluxes computed for model stellar atmospheres .
the stellar angular diameters obtained from fitting the fluxes at 3.5 @xmath19 m are in good agreement with observational values and with angular diameters deduced from the relation between visual surface brightness and ( @xmath296 ) colour .
the temperatures obtained from the shape of the flux curves are in satisfactory agreement with other temperature estimates .
since the average error is expected to be well within 10% , the error for the angular diameter is estimated to be in the order of 5% .
smith & lambert ( 1990 ) have determined the chemical composition of a sample of m , ms and s giants . the use of a slightly different set of lines for the molecular vibra - rotational lines , along with improved @xmath336-values for cn and nh , lead to a small difference in the carbon , nitrogen and oxygen abundance compared to smith & lambert ( 1985 ) .
kjrgaard et al . ( 1982 ) examined two cases for which they derived equations for the approximate absorption coefficient @xmath337 . in case ( a ) most of the carbon and nitrogen
are still considered to be free atoms , assumed for t@xmath1 @xmath338 5000 k , while in case ( b ) almost all the c and n are in the form of co and n@xmath117 , assumed for t@xmath1 @xmath276 4500 k. looking at the relatively high partial pressure of cn , it seems that the assumption of complete association of n into n@xmath117 is not valid for our giant models . attributing all depletion to the formation of n@xmath117 seems to be more valid in a dwarf model ( bell & tripicco 1991 ) .
the formula can however still give a qualitative view of the influence of the different parameters .
for case ( b ) kjrgaard et al .
deduced an approach for cn : @xmath339 using the approach of kjrgaard et al .
, similar equations for co , sio , oh , h@xmath117o and nh are derived . for co
, we have @xmath340 in the assumption that all carbon is locked into co , + @xmath119(co ) may also be written as @xmath341 with the approximation of n(e ) and n(h)/n(e ) in the line - forming region being ( kjrgaard et al . , 1982 )
@xmath342 the approximate absorption coefficient of sio is defined as @xmath343 for oxygen - rich giants , n(o i ) may be approximated by @xmath344 the equilibrium of silicon is dominated by sio and si i ( si ii is present in the deeper layers as well ) .
for example , in the model with t@xmath1= 4050 k , @xmath3 g = 1.00 , z = 0.00 , m = 1.5 m@xmath6 , @xmath8 = 2.0 @xmath10 , the proportion of the number densities are given in table [ tablesio ] .
since the ratio @xmath345 changes rapidly in the outer layers of the photosphere , a rough approximation to @xmath346 , valid in the region where @xmath347 , is then + @xmath348 so that @xmath349 the approximate absorption coefficient of nh is written as @xmath358 the equilibrium of the nitrogen species is dominated by the n@xmath117 formation .
so , @xmath359 with @xmath360 resulting in @xmath361^{1/2}}.\ ] ] consequently , @xmath362 ld acknowledges support from the science foundation of flanders .
bp thanks s. johansson and the staff of the atomic spectroscopy division at lund university for their hospitality during part of this work .
this research has made use of the simbad database , operated at cds , strasbourg , france .
anders e. , grevesse n. , 1989 , geochimica et cosmochimica acta 53 , 197 aoki w. , tsuji t. , 1997 , a&a 328 , 175 barnes t.g . , evans d.s . , 1976 , mnras 174 , 489 baschek b. , scholz m. , wehrse r. , 1991 , a&a 246 , 374 bell r.a . , gustafsson b. , 1989 , mnras 236 , 653 bell r.a . , trippico m.j . , 1991 , aj 102 , 777 bessell m.s . , castelli f. , plez b. , 1998 , a&a 333 , 231 blackwell d.e . , shallis m.j . , 1977 ,
mnras 180 , 177 blackwell d.e . ,
petford a.d . , arribas s. , haddock d.j . , selby m.j . , 1990 ,
a&a 232 , 396 blackwell d.e .
, lynas - gray a.e . ,
petford a.d . , 1991 , a&a 245 , 567 bonnell j.t . , bell r.a . , 1993 ,
mnras 264 , 319 brockwell p.j . , davis r.a . , 1991 ,
time series : theory and methods , springer - verlag , new york , inc .
, 2@xmath363 edition brown j.a . , tomkin j. , lambert d.l . , 1983 , apj 265 , l93 burnashev v.i . , 1983 ,
krymskaia astrofiz .
67 , 13 charbonnel c. , 1994 , a&a 1994 , 282 claret a. , gimnez a. , 1995 , a&as 114 , 549 cohen m. , walker r.g .
, witteborn f.c . , 1992 , aj 104 , 203 cohen m. , witteborn f.c .
, bregman j.d . , wooden d.h . , salama a. , metcalfe l. , 1996a , aj 112 , 241 cohen m. , witteborn f.c . , carbon d.f . , davies j.k . , wooden d.h . ,
bregman j.d . , 1996b , aj 112 , 2274 conti p.s . , greenstein j.l . ,
spinard h. , wallerstein g. , vardya m.s .
, 1967 , apj 148 , 105 currie d.g . ,
knapp s.l . ,
liewer k.m .
, 1974 apj 187 , 131 davis j. , 1997 , in : bedding t.r . ,
booth a.j . , davis j. ( eds ) fundamental stellar properties : the interaction between observation and theory , iau symp .
189 , p. 31
decin l. , 2000 , synthetic spectra of cool stars observed with the iso short - wavelength spectrometer : improving the models and the calibration of the instrument , phd .
thesis , university of leuven decin l. , cohen m. , eriksson k. , et al .
, 1997 , in : heras a.m. , leech k. , trams n.r . , perry , m. ( eds . ) the first iso workshop on analytical spectroscopy , esa sp-149 , p. 185
de graauw th .
, haser l.n . ,
beintema d.a .
, et al . , 1996 , a&a 315 , l49 di benedetto g.p . , 1998 , a&a 339 , 858 di benedetto g.p . , rabbia y. , 1987 , a&a 188 , 114 edlen b. , 1966 , metrologia 2 , 71 el eid m.f .
, 1994 , a&a 285 , 915 engelke c.w . , 1992 , aj 104 , 1248 fasano g. , franceschini a. , 1987 , mnras 225 , 155 fernndez - villacaas j.l . , reggo m. , cornide m. , 1990 , aj 99 , 1961 fracassini m. , pasinetti - fracassini l.e . , pastori l. , pironi r. , 1988 , bull .
cds 35 , 121 geller m. , sauval a.j .
, grevesse n. , farmer c.b . ,
norton r.h . , 1991 ,
a&a 249 , 550 gliese w. , 1969 , verffentlichungen des astronomishen rechen - instituts heidelberg 22,1 goldman a. , schoenfeld w.g . , goorvitch d. , et al . , 1998 , j. quant .
transfer 59 , 453 goorvitch d. , 1994 , apjs 95 , 535 gosset e. , 1987 , a&a 188 , 258 gray d.f .
, 1992 , `` the observation and analysis of stellar photospheres '' , cambridge astrophysics series , cambridge : cambridge university press , 2nd ed .
, isbn 0521403200 . grevesse n. , lambert d.l .
, sauval a.j . , et al . , 1990 , a&a 32 , 225 grevesse n. , lambert d.l . , sauval a.j . ,
et al . , 1991 , a&a 242 , 488 gustafsson b. , bell r.a . , 1979 , a&a 74 , 313 gustafsson b. , bell r.a . , eriksson k. , nordlund , 1975 , a&a 42 , 407 hjek j. , 1969 , `` nonparametric statistic '' , lehmann e.l .
, holden - day , inc .
harris m.j . ,
lambert d.l .
, 1984 , apj 285 , 674 harris m.j .
, lambert d.l . ,
smith v.v . , 1988 , apj 325 , 768 henyey l. , vardya m.s . , bodenheimer p. , 1965
, aj 142 , 841 hirata r. , horaguchi t. , 1995 , internet : + http://cdsweb.u-strasbg.fr/htbin/cat?vi/69 hoffleit d. , jaschek c. , 1982 , `` the bright star catalogue '' , new haven : yale university observatory , 4th ed .
johnson h.l , 1966 , ara&a 4 , 193 jrgensen u.g . , 1994 , in : jrgensen u.g .
( ed . ) dominating molecules in the photospheres of cool stars , iau symp .
146 , p. 29
kjrgaard p. , gustafsson b. , walker g.a.h .
, hultqvist l. , 1982 , a&a 115 , 145 kovcs n. , 1983 , a&a 120 , 21 kurucz r.l . , 1993 ,
cd - rom set , cambridge , ma : smithsonian astrophysical observatory .
lambert d.l . , 1976 ,
msrsl 9 , 405 lambert d.l . , ries l.m . , 1981 ,
apj 248 , 228 lambert d.l . , dominy j.f . , sivertsen s. , 1980 , apj 235 , 114 langhoff s.r . , bauschlicher c.w.jr
, 1993 , chem .
letters 211 , 305 linsky j.l , ayres t.r . , 1978 ,
apj 220 , 619 lorente r. , 1998 , spectral resolution of sws aot 1 , iso - sws online documentation luck r.e .
, challener s.l . , 1995 , aj 110 , 2968 manduca a. , 1979 , a&as 36 , 411 manduca a. , bell r.a . , gustafsson b. , 1981 , apj 243 , 883 mcwilliam a. , 1990 , apjs 74 , 1075 melen f. , grevesse n. , sauval a.j . , et al . , 1989 ,
j. mol .
spectrosc .
134 , 305 melen f. , sauval a.j . , grevesse n. , et al . , 1995 ,
j. mol .
spectrosc .
174 , 490 mozurkewich d. , johnston k.j . ,
simon r.s . , bowers p.f .
, gaume r. , 1991 , aj 101 , 2207 nordlund , dravins d. , 1990 , a&a 228 , 155 partridge h. , schwenke d.w .
, 1997 , j. chem .
106 , 4618 perrin g. , coud du foresto v. , ridgway s.t . , et al . , 1998 , a&a 331 , 619 plez b. , brett j.m .
, nordlund , 1992 , a&a 256 , 551 plez b. , smith .
v.v . , lambert d.l . , 1993 , aj 418 , 812 pratt j.w . , gibbons j.d . , 1981 , `` concepts of nonparametric
theory '' , springer - verlag , new york inc .
quirrenbach a. , mozurkewich d. , armstrong j.t . ,
buscher d.f . ,
hummel c.a . , 1993 ,
apj 406 , 215 ridgway s.t . ,
joyce r.r . ,
white n.m . , wing r.f .
, 1980 , apj 235 , 126 ridgway s.t . ,
jacoby g.h . ,
joyce r.r .
, siegel m.j . , wells d.c . , 1982 , aj 87 , 1044 ries l.m . , 1981 , ph.d .
thesis texas univ .
, arlington scargle j.d . ,
strecker d.w . , 1979 ,
apj 228 , 838 schaeidt s.g . ,
morris p.w .
, salama a. , et al . , 1996 , a&a 315 , l55 scholz m. , 1998 , in : bedding t.r . , booth a.j . , davis j. ( eds ) fundamental stellar properties : the interaction between observation and theory , iau symp .
189 , p. 51
schwenke d. , 1997 , internet : + http://george.arc.nasa.gov/~dschwenke smith v.v .
, lambert d.l . , 1985 ,
apj 294 , 326 smith v.v . , lambert d.l .
, 1990 , apjs 72 , 387 taylor b.j . , 1999 , a&as 134 , 523 tomkin j. , lambert d.l .
, 1974 , apj 193 , 631 tomkin j. , lambert d.l . , luck r.e .
, 1975 , apj 199 , 436 tomkin j. , luck r.e . ,
lambert d.l . , 1976 , apj 210 , 694 trimble v. , bell r.a . , 1981 , quart .
22 , 361 tsuji t. , 1981 , a&a 99 , 48 tsuji t. , 1986 , a&a 156 , 8 tsuji t. , 1991 , a&a 245 , 203 valentijn e.a . , feuchtgruber h. , kester d.j.m . , et al .
, 1996 , a&a 315 , l60 vandenbussche b. , 1999 , the iso - sws relative spectral response calibration , iso - sws online documentation van der bliek n.s .
, morris p.w . , vandenbussche b. , et al . , 1998 , in : bedding t.r . ,
booth a.j . , davis j. ( eds . ) fundamental stellar properties : the interaction between observation and theory , iau symp . 189 , p. 89
van malderen r. , kester d.j.m .
, decin l. , vandenbussche b. , waelkens c. , 1999 , in : hron j. , hfner s.(eds . )
abstracts of the 2nd austrian iso workshop : atmospheres of m , s and c giants , p. 79
van paradijs j. , meurs e.j.a .
, 1974 , a&a 35 , 225 volk k. , cohen m. , 1989 , aj 98 , 1918 wilson p.m. , 1972 , mnras 158 , 361 wilson o.c . , 1976 ,
apj 205 , 823 witteborn f.c . , cohen m. , bregman j.d . , et al . , 1999 , aj 117 , 2552 zahn j.p
. , 1992 , a&a 265 , 115 | a detailed spectroscopic study of the iso - sws data of the red giant @xmath0 tau is presented , which enables not only the accurate determination of the stellar parameters of @xmath0 tau , but also serves as a critical review of the iso - sws calibration .
this study is situated in a broader context of an iterative process in which both accurate observations of stellar templates and cool star atmosphere models are involved to improve the iso - sws calibration process as well as the theoretical modelling of stellar atmospheres .
therefore a sample of cool stars , covering the whole a0
m8 spectral classification , has been observed in order to disentangle calibration problems and problems in generating the theoretical models and corresponding synthetic spectrum . by using stellar parameters found in the literature
large discrepancies were seen between the iso - sws data and the generated synthetic spectrum of @xmath0 tau .
a study of the influence of various stellar parameters on the theoretical models and synthetic spectra , in conjunction with the kolmogorov - smirnov test to evaluate objectively the goodness - of - fit , enables us to pin down the stellar parameters with a high accuracy : t@xmath1 @xmath2 k , @xmath3 g@xmath4 , m @xmath5 m@xmath6 , z @xmath7 dex , @xmath8 @xmath9 @xmath10 , @xmath11 @xmath12 , @xmath13(c ) @xmath14 dex , @xmath13(n ) @xmath15 dex , @xmath13(o ) @xmath16 dex and @xmath17 @xmath18 mas .
these atmospheric parameters were then compared with the results provided by other authors using other methods and/or spectra . |
the ability to standardize type ia supernova ( sn ia ) light curves yields precise cosmological distance indicators ( e.g. , * ? ? ? * ) . despite these empirical relationships ,
questions remain about the underlying physics and progenitor systems .
the sn itself is generally accepted to be the thermonuclear explosion of a c / o white dwarf ( wd ) , but the nature of the companion star remains ambiguous .
evidence now exists for both single - degenerate and double - degenerate channels ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
recent studies reveal a subsample of sne ia that exhibit signs of significant interaction between the forward shock and a dense circumstellar medium ( hereafter sne ia - csm ; @xcite , and references within ) .
the spectra show many similarities to those of sne iin , defined by their strong and relatively narrow hydrogen emission lines generated from a slow , dense , pre - existing csm ( @xcite ; see @xcite for a review ) . the sne ia - csm also have slightly larger peak luminosities than typical sne ia , in the range @xmath2 mag .
detailed observations of the nearby sn ia - csm ptf11kx confirm a probable single - degenerate channel for at least this particular event @xcite , although mass loss from a violent prompt merger after a common envelope ( i.e. , a core - degenerate scenario ) has also been proposed @xcite .
two of the most well - studied sne ia - csm are sne 2002ic and 2005gj @xcite .
the data suggest substantial mass loss ( of order @xmath3 m@xmath4 yr@xmath5 ) from the companion star and significant amounts of warm dust emitting in the near - infrared ( near - ir ) .
these characteristics are also similar to those of sne iin . owing to their dense csm , sne iin exhibit late - time ( @xmath6 day ) ir emission from warm dust that is continuously heated by visible and x - ray radiation generated by ongoing csm interaction ( e.g. , *
* ; * ? ? ?
the csm geometry derived from these dust shells reveals important clues about the progenitor mass - loss history . in this _ letter _
we present unpublished archival @xmath0 data on sne 2002ic and 2005gj obtained @xmath1 and 500 days post - discovery , respectively .
section [ sec:2 ] lists the details of the observations ; @xmath7 photometry constrains the dust mass and temperature , and thus the luminosity .
we explore the origin and heating mechanism of the dust in [ sec:3 ] .
section [ sec:4 ] presents our conclusions .
l c c c c c c c c c c 2002ic & 3386 & 795 & 3248 & 10550272 & 280 & 4000 & 0.150(0.015 ) & 0.136(0.012 ) & 0.092(0.008 ) & 0.052(0.004 ) + 2002ic & 3770 & 1179 & 20256 & 14455040 & 280 & 3600 & 0.060(0.009 ) & 0.063(0.008 ) & 0.052(0.006 ) & 0.040(0.004 ) + 2002ic & 3961 & 1370 & 30292 & 17965824 & 280 & 2400 & 0.042(0.008 ) & 0.045(0.007 ) & 0.037(0.005 ) & 0.031(0.003 ) + 2002ic & 4356 & 1765 & 40619 & 23107840 & 280 & 2400 & @xmath80.03 & @xmath80.03 & @xmath80.03 & @xmath80.03 + 2005gj & 3778 & 139 & 264 & 16868096 & 268 & 2400 & 0.130(0.014 ) & 0.092(0.010 ) & 0.036(0.005 ) & 0.017(0.003 ) + 2005gj & 4004 & 365 & 30733 & 19308800 & 268 & 2400 & 0.183(0.016 ) & 0.110(0.011 ) & 0.049(0.006 ) & 0.024(0.003 ) + 2005gj & 4149 & 510 & 30733 & 19309056 & 268 & 2400 & 0.276(0.020 ) & 0.199(0.014 ) & 0.113(0.009 ) & 0.049(0.004 ) a search in the _ spitzer _ heritage archive ( sha ) revealed unpublished observations of sne 2002ic and 2005gj , summarized in table [ tab1 ] .
the sha provide access to the post basic calibrated data ( pbcd ) , which are already fully coadded and calibrated .
figure [ f1 ] shows false - color images of the combined 3.6 , 4.5 , and 5.8 channel images at a single epoch .
the sn host galaxies tend to be bright and exhibit background - flux variations on rapid spatial scales .
template subtraction can reduce photometric confusion from the underlying galaxy , but no pre - sn _ spitzer _ templates exist for these galaxies .
we therefore performed aperture photometry with daophot / apphot in iraf .
sn 2005gj falls @xmath9 from the galactic nucleus @xcite . to minimize contributions from the underlying galaxy in both cases ,
a 2-pixel radius was chosen and aperture corrections were applied , although removal of nuclear contributions is admittedly difficult .
+ + assuming thermal emission to be the dominant flux component , the mid - ir photometry probes the peak of the blackbody produced by warm grains .
the flux can be fit as a function of the dust temperature , @xmath10 , and mass , @xmath11 , @xmath12 assuming optically thin dust with particle radius @xmath13 , at a distance @xmath14 from the observer , thermally emitting at a single equilibrium temperature ( e.g. , * ? ? ?
* ) , where @xmath15 is the planck blackbody function and @xmath16 is the dust absorption coefficient . for simplicity , we assume a simple dust population of a single size composed entirely of amorphous carbon ( ac ) .
figure [ f2 ] shows the best fit of equation [ eqn : flux2 ] obtained with the idl mpfit function @xcite , which minimizes the value of @xmath17 by varying @xmath11 and @xmath10 . with only four photometry points at each epoch ,
we limit our fits to a single component ( see figure [ f2 ] ) .
table [ tab2 ] lists the best - fit parameters for ac grains of size @xmath18 .
l c c c c [ tab_phot ] 2002ic & 795 & 0.012 & 590 & 1.64@xmath19 + 2002ic & 1179 & 0.016 & 495 & 8.70@xmath20 + 2002ic & 1370 & 0.014 & 481 & 6.34@xmath20 + 2005gj & 139 & 0.001 & 845 & 1.06@xmath19 + 2005gj & 365 & 0.002 & 847 & 1.55@xmath19 + 2005gj & 510 & 0.006 & 725 & 2.34@xmath19
figure [ f3 ] plots the corresponding ir luminosity evolution for each object .
sn 2002ic remains bright ( @xmath21 l@xmath4 ) for more than 2 yr post - discovery , but continues to fade throughout the observations .
by contrast , sn 2005gj brightens after 1 year post - discovery .
while we only have mid - ir photometry available at these epochs , we explore the constraints these data provide on the circumstellar environment .
the source of the mid - ir emission is warm dust , but the origin and heating mechanism of the dust are less clear
. the dust may be either newly formed or pre - existing , and either shock or radiatively heated ; see @xcite for a full discussion . to discriminate between possible scenarios , we assume a spherically symmetric , optically thin dust shell and calculate the blackbody radius , @xmath22^{1/2}$ ] , which sets a _ minimum _ shell size .
a luminosity @xmath23 l@xmath4 and dust temperature @xmath24 k yield a blackbody radius of @xmath25 cm .
this radius is larger than that of a forward shock moving at @xmath26 10,000 on day 800 .
this point seemingly rules out , particularly for the younger sn 2005gj , the possibility that the majority of the observed dust condensed in either the more slowly moving ejecta or the cold , dense shell that can form behind the forward shock .
[ we note , however , that @xcite observed evidence of some newly formed dust in sn 2005gj via increasing absorption in the red h@xmath27 wing .
while they do not estimate the dust mass , the @xmath7 observations suggest that this dust does not contribute significantly to the overall mid - ir flux . ]
the alternative to newly formed dust is a pre - existing shell .
again , the fact that the blackbody radius is beyond the forward - shock radius rules out the likelihood of shock heating .
furthermore , an ir light - echo scenario ( e.g. , * ? ? ?
* ) , in which the dust shell is heated by the peak sn luminosity , is unfeasible .
the implied shell radii ( @xmath28 ) would require peak luminosities of nearly @xmath29 l@xmath4 to heat the dust to the observed temperatures .
in fact , the observed dust - shell parameters ( i.e. , radius , temperature , mass ) are comparable to those seen in sne iin ( e.g. , * ? ? ? * ; * ? ? ?
* ) . in these cases , ongoing interaction between the forward shock and dense csm
produce optical , uv , and x - ray emission that radiatively heat a pre - existing dust shell .
assuming an optically thin dust shell , the observed dust temperature ( @xmath10 ) and shell radius ( @xmath30 ) require a combined optical , ultraviolet , and/or x - ray flux given by @xmath31 for a dust bulk ( volume ) density @xmath32 and an effective sn blackbody temperature @xmath33 , where @xmath34 is the planck blackbody function , @xmath35 is the dust absorption efficiency , and @xmath36 is the dust absorption coefficient .
figure 4 in @xcite shows that the blackbody radii of sne 2002ic and 2005gj [ @xmath37 ( 0.51.0 ) @xmath38 cm ] at temperatures @xmath24750 k require optical and/or x - ray luminosities in the range @xmath39 l@xmath4 . while we do not have optical or x - ray observations at these epochs , the most recent optical observations of sne 2002ic and 2005gj are consistent , with measured luminosities of @xmath40 and @xmath41 l@xmath4 on days 250 and 149 , respectively @xcite .
mid - ir wavelengths probe the characteristics of the csm at the dust - shell radius .
assuming a dust - to - gas mass ratio expected in the h - rich envelope of a massive star , @xmath42 , the dust - shell mass can be tied to the progenitor s total mass - loss rate , @xmath43 for a progenitor wind speed @xmath44 .
the relatively narrow lines observed in sne ia - csm originate in the slow pre - shocked csm and can be used to approximate the progenitor wind speed . the precursor wind velocities for sne 2002ic and 2005gj are @xmath45 100 and 60 , respectively @xcite . assuming a thin shell , @xmath46 , wind speed @xmath45 60 , and radius @xmath47 cm , the approximate mass - loss rate to produce the observed dust shell is @xmath48 m@xmath4 yr@xmath5 .
a smaller dust - shell radius would require an even larger mass - loss rate .
the optical and/or x - ray emission generated by csm interaction traces the mass loss at the inner radii . assuming a density @xmath49 wind profile , the rate can be written as a function of the optical / x - ray luminosity , progenitor wind speed , and shock velocity ( e.g. , * ? ? ?
* ; * ? ? ?
* ) : @xmath50 where @xmath51 is the efficiency of converting shock kinetic energy into visual light .
we assume a value @xmath52 , although the conversion efficiency can vary with wind density and shock speed .
again , we do not have late - time optical or x - ray observations , but we do have theoretical estimates from equation [ eqn : lbol ] in [ sec : origin ] .
an optical luminosity @xmath53 l@xmath4 , wind speed @xmath45 120 , shock velocity @xmath54 10,000 @xcite , and conversion efficiency @xmath55 correspond to a mass - loss rate @xmath56 m@xmath4 yr@xmath5 . while the variables used above are only order - of - magnitude approximations , the derived rate for @xmath57 reveals two things about the circumstellar medium .
( 1 ) the difference between the mass - loss rates ( @xmath58 ) suggests that the dust shells were formed during a period of increased , nonsteady mass loss .
( 2 ) compared to mass - loss rates derived from optical data at earlier epochs , our estimate of @xmath56 m@xmath4 yr@xmath5 is consistent with sn 2002ic on day 250 ( for @xmath59 ; @xcite ) , but at least an order of magnitude larger than that measured for sn 2005gj on day 74 @xcite .
the increased circumstellar density derived for sn 2005gj at this late time suggests the presence of another shell of material .
the decline of sn 2002ic occurs at @xmath60800 days , corresponding to the time at which the forward shock ( @xmath54 10,000 ) would reach the blackbody radius ( @xmath25 cm ) . the declining light curve may be attributed to the forward shock overtaking and destroying the dust and/or a decreasing amount of csm interaction accompanied by a declining radiative heating source .
alternatively , the rebrightening of the pre - existing dust in sn 2005gj is likely due to radiative heating by renewed shock interaction with this dense circumstellar shell . from equation [ eqn : inner ] , an order - of - magnitude increase in mass loss results in an increase in the optical luminosity by a factor of 34 , assuming the shock velocity decreases to 0.8 of its former value .
for a constant blackbody radius of @xmath61 cm , equation [ eqn : lbol ] and figure 10 in @xcite show that an increase in the optical luminosity from @xmath62 to @xmath63 l@xmath4 results in a dust temperature increase from @xmath64 to 750 k. since @xmath65 , this change in temperature results in a luminosity increase of a factor of 2.4 .
while we observe a luminosity increase of this magnitude ( see table [ tab2 ] ) , the inferred dust temperature actually _ decreases _ by @xmath66 k.
the implication is that the blackbody radius must also increase from @xmath67 to @xmath68 cm , which would explain a lower dust temperature along with a higher dust mass and luminosity .
the problem with invoking this scenario is that the only way to increase the blackbody radius ( assuming a spherically symmetric shell of dust ) would be for the increased optical luminosity to vaporize all dust out to the new blackbody radius . for a dust vaporization temperature of @xmath69
k , equation [ eqn : lbol ] and figure 8 of @xcite show that to vaporize dust out to @xmath68 cm requires a luminosity of @xmath70 l@xmath4 , which is not likely .
the more probable explanation is that with only four data points , the slopes of the curves in figure [ f2 ] and , therefore the temperatures , are biased by contamination from the underlying galaxy nucleus ( @xmath9 away and a 0.6 pixel scale ) .
the total dust luminosity , which is less sensitive to the slope , is most consistent with radiative heating by renewed shock interaction .
this _ letter _ presents unpublished archival @xmath7 data on sne ia - csm 2002ic and 2005gj obtained @xmath1 and 500 days post - discovery .
the mid - ir data show evidence of emission from warm dust . while we do not have simultaneous observations at shorter wavelengths , the warm - dust parameters are most constant with a pre - existing dust shell heated by a combination of optical , uv , and x - ray emission continuously generated by ongoing csm interaction .
the degree of csm interaction dictates the dust temperature and , thereby , the luminosity . in the case of
sn 2005gj , the mid - ir luminosity nearly doubles more than 1 year post - discovery , suggesting an increasing amount of csm interaction .
we attribute this renewed shock interaction to a dense circumstellar shell produced during a period of increased mass loss by the progenitor companion .
while progenitor mass loss suggests a single - degenerate channel , such rates have also been derived for both the core - degenerate and double - degenerate models @xcite .
future multi - wavelength observations of sne ia - csm will be necessary to better trace the complete mass - loss history and constrain the nature of the companion star . + this work is based on archival data obtained with the _ spitzer space telescope _ , which is operated by the jet propulsion laboratory , california institute of technology , under a contract with nasa .
support for this work was provided by nasa through an award issued by jpl / caltech ( p90031 ) .
we also acknowledge generous financial assistance from the christopher r. redlich fund , the tabasgo foundation , and nsf grant ast-1211916 . | a growing number of observations reveal a subset of type ia supernovae undergoing circumstellar interaction ( sne ia - csm ) .
we present unpublished archival @xmath0 data on sne ia - csm 2002ic and 2005gj obtained @xmath1 and 500 days post - discovery , respectively .
both sne show evidence for late - time mid - infrared ( mid - ir ) emission from warm dust .
the dust parameters are most consistent with a pre - existing dust shell that lies beyond the forward - shock radius , most likely radiatively heated by optical and x - ray emission continuously generated by late - time csm interaction . in the case of sn 2005gj , the mid - ir luminosity more than doubles after 1 year post - discovery . while we are not aware of any late - time optical - wavelength observations at these epochs , we attribute this rebrightening to renewed shock interaction with a dense circumstellar shell . + |
the availability of ubiquitous communication through the internet is driving the migration of business transactions from direct contact between people to electronically mediated interactions .
people interact electronically either through human - computer interfaces or through programs representing humans , so - called agents . in either case ,
no physical interactions among entities occur , and the systems are much more susceptible to fraud and deception . traditional methods to avoid cheating involve cryptographic schemes and _ trusted third parties _ ( ttp s ) that overlook every transaction .
such systems are very costly , introduce potential bottlenecks , and may be difficult to deploy due to the complexity and heterogeneity of the environment : e.g. , agents in different geographical locations may be subject to different legislation , or different interaction protocols .
reputation mechanisms offer a novel and effective way of ensuring the necessary level of trust which is essential to the functioning of any market .
they are based on the observation that agent strategies change when we consider that interactions are repeated : the other party remembers past cheating , and changes its terms of business accordingly in the future . therefore , the expected gains due to future transactions in which the agent has a higher reputation can offset the loss incurred by not cheating in the present .
this effect can be amplified considerably when such reputation information is shared among a large population , and thus multiplies the expected future gains made accessible by honest behavior .
existing reputation mechanisms enjoy huge success .
systems such as ebay or amazon implement reputation mechanisms which are partly credited for the businesses success .
studies show that human users seriously take into account the reputation of the seller when placing bids in online auctions @xcite , and that despite the incentive to free ride , feedback is provided in more than half of the transactions on ebay @xcite .
one important challenge associated with designing reputation mechanisms is to ensure that truthful feedback is obtained about the actual interactions , a property called _ incentive - compatibility_. rational users can regard the private information they have observed as a valuable asset , not to be freely shared .
worse even , agents can have external incentives to misreport and thus manipulate the reputation information available to other agents @xcite . without proper measures ,
the reputation mechanism will obtain unreliable information , biased by the strategic interests of the reporters .
honest reporting incentives should be addressed differently depending on the predominant role of the reputation mechanisms .
the _ signaling _ role is useful in environments where the service offered by different providers may have different quality , but all clients interacting with the same provider are treated equally ( markets with _ adverse selection _ ) .
this is the case , for example , in a market of web - services .
different providers possess different hardware resources and employ different algorithms ; this makes certain web - services better than others .
nevertheless , all requests issued to the same web - service are treated by the same program .
some clients might experience worse service than others , but these differences are random , and not determined by the provider .
the feedback from previous clients statistically estimates the quality delivered by a provider in the future , and hence signals to future clients which provider should be selected .
the _ sanctioning _ role , on the other hand , is present in settings where service requests issued by clients must be individually addressed by the provider
. think of a barber , who must skillfully shave every client that walks in his shop .
the problem here is that providers must exert care ( and costly effort ) for satisfying every service request
. good quality can result only when enough effort was exerted , but the provider is better off by exerting less effort : e.g. , clients will anyway pay for the shave , so the barber is better off by doing a sloppy job as fast as possible in order to have time for more customers .
this _ moral hazard _
situation can be eliminated by a reputation mechanism that punishes providers for not exerting effort .
low effort results in negative feedback that decreases the reputation , and hence the future business opportunities of the provider .
the future loss due to a bad reputation offsets the momentary gain obtained by cheating , and makes cooperative behavior profitable .
there are well known solutions for providing honest reporting incentives for signaling reputation mechanisms . since all clients interacting with a service receive the same quality ( in a statistical sense ) , a client s private observation influences her belief regarding the experience of other clients . in the web - services market
mentioned before , the fact that one client had a bad experience with a certain web - service makes her more likely to believe that other clients will also encounter problems with that same web - service .
this correlation between the client s private belief and the feedback reported by other clients can be used to design feedback payments that make honesty a nash equilibrium . when submitting feedback , clients get paid an amount that depends both on the the value they reported and on the reports submitted by other clients .
as long as others report truthfully , the expected payment of every client is maximized by the honest report thus the equilibrium . and show that incentive - compatible payments can be designed to offset both reporting costs and lying incentives . for _ sanctioning _ reputation mechanisms the same payment schemes are not guaranteed to be incentive - compatible .
different clients may experience different service quality because the provider decided to exert different effort levels .
the private beliefs of the reporter may no longer be correlated to the feedback of other clients , and therefore , the statistical properties exploited by are no longer present . as an alternative , we propose different incentives to motivate honest reporting based on the repeated presence of the client in the market .
game theoretic results ( i.e. , the _ folk theorems _ ) show that repeated interactions support new equilibria where present deviations are made unattractive by future penalties . even without a reputation mechanism ,
a client can guide her future play depending on the experience of previous interactions . as a first result of this paper ,
we describe a mechanism that indeed supports a cooperative equilibrium where providers exert effort all the time .
the reputation mechanism correctly records when the client received low quality .
there are certainly some applications where clients repeatedly interact with the same seller with a potential moral hazard problem .
the barber shop mentioned above is one example , as most people prefer going to the same barber ( or hairdresser ) .
another example is a market of delivery services .
every package must be scheduled for timely delivery , and this involves a cost for the provider .
some of this cost may be saved by occasionally dropping a package , hence the moral hazard .
moreover , business clients typically rely on the same carrier to dispatch their documents or merchandise .
as their own business depends on the quality and timeliness of the delivery , they do have the incentive to form a lasting relationship and get good service .
yet another example is that of a business person who repeatedly travels to an offshore client .
the business person has a direct interest to repeatedly obtain good service from the hotel which is closest to the client s offices .
we assume that the quality observed by the clients is also influenced by environmental factors outside the control of , however observable by , the provider . despite the barber s best effort
, a sudden movement of the client can always generate an accidental cut that will make the client unhappy .
likewise , the delivery company may occasionally lose or damage some packages due to transportation accidents .
nevertheless , the delivery company ( like the barber ) eventually learns with certainty about any delays , damages or losses that entitle clients to complain about unsatisfactory service .
the mechanism we propose is quite simple . before asking feedback from the client ,
the mechanism gives the provider the opportunity to acknowledge failure , and reimburse the client .
only when the provider claims good service does the reputation mechanism record the feedback of the client .
contradictory reports ( the provider claims good service , but the client submits negative feedback ) may only appear when one of the parties is lying , and therefore , both the client and the provider are sanctioned : the provider suffers a loss as a consequence of the negative report , while the client is given a small fine .
one equilibrium of the mechanism is when providers always do their best to deliver the promised quality , and truthfully acknowledge the failures caused by the environmental factors .
their `` honest '' behavior is motivated by the threat that any mistake will drive the unsatisfied client away from the market .
when future transactions generate sufficient revenue , the provider does not afford to risk losing a client , hence the equilibrium .
unfortunately , this socially desired equilibrium is not unique .
clients can occasionally accept bad service and keep returning to the same provider because they do nt have better alternatives .
moreover , since complaining for bad service is sanctioned by the reputation mechanism , clients might be reluctant to report negative feedback .
penalties for negative reports and the clients lack of choice drives the provider to occasionally cheat in order to increase his revenue . as a second result
, we characterize the set of pareto - optimal equilibria of our mechanism and prove that the amount of unreported cheating that can occur is limited by two factors .
the first factor limits the amount of cheating in general , and is given by the quality of the alternatives available to the clients .
better alternatives increase the expectations of the clients , therefore the provider must cheat less in order to keep his customers .
the second factor limits the amount of unreported cheating , and represents the cost incurred by clients to establish a reputation for reporting the truth . by stubbornly exposing bad service when it happens , despite the fine imposed by the reputation mechanism , the client signals to the provider that she is committed to always report the truth .
such signals will eventually change the strategy of the provider to full cooperation , who will avoid the punishment for negative feedback . having a reputation for reporting truthfully is of course , valuable to the client ; therefore , a rational client accepts to lie ( and give up the reputation ) only when the cost of building a reputation for reporting honestly is greater than the occasional loss created by tolerated cheating .
this cost is given by the ease with which the provider switches to cooperative play , and by the magnitude of the fine imposed for negative feedback .
concretely , this paper proceeds as follows . in section [ related_work ]
we describe related work , followed by a more detailed description of our setting in section [ setting ] .
section [ gtanalysis ] presents a game theoretic model of our mechanism and an analysis of reporting incentives and equilibria .
here we establish the existence of the cooperative equilibrium , and derive un upper bound on the amount of cheating that can occur in any pareto - optimal equilibrium .
in section [ buidingreputation ] we establish the cost of building a reputation for reporting honestly , and hence compute an upper bound on the percentage of false reports recorded by the reputation mechanism in any equilibrium .
we continue in section [ evilbuyers ] by analyzing the impact of malicious buyers that explicitly try to destroy the reputation of the provider .
we give some initial approximations on the worst case damage such buyers can cause to providers .
further discussions , open issues and directions for future work are discussed in section [ future_work ] .
finally , section [ conclusions ] concludes our work .
the notion of _ reputation _ is often used in game theory to signal the commitment of a player towards a fixed strategy .
this is what we mean by saying that _ clients establish a reputation for reporting the truth _ : they commit to always report the truth . building a reputation
usually requires some incomplete information repeated game , and can significantly impact the set of equilibrium points of the game .
this is commonly referred to as the _ reputation effect _ ,
first characterized by the seminal papers of , and .
the reputation effect can be extended to all games where a player ( @xmath0 ) could benefit from committing to a certain strategy @xmath1 that is not credible in a complete information game : e.g. , a monopolist seller would like to commit to fight all potential entrants in a chain - store game @xcite , however , this commitment is not credible due to the cost of fighting . in an incomplete information game where the commitment type has positive probability , @xmath0 s opponent ( @xmath2 ) can at some point become convinced that @xmath0 is playing as if she were the commitment type .
at that point , @xmath2 will play a best response against @xmath1 , which gives @xmath0 the desired payoff .
establishing a reputation for the commitment strategy requires time and cost .
when the higher future payoffs offset the cost of building reputation , the reputation effect prescribes minimum payoffs any equilibrium strategy should give to player @xmath0 ( otherwise , @xmath0 can profitably deviate by playing as if she were a commitment type ) . study the class of all repeated games in which a long - run player faces a sequence of single - shot opponents who can observe all previous games .
if the long - run player is sufficiently patient and the single - shot players have a positive prior belief that the long - run player might be a commitment type , the authors derive a lower bound on the payoff received by the long - run player in any nash equilibrium of the repeated game .
this result holds for both finitely and infinitely repeated games , and is robust against further perturbations of the information structure ( i.e. , it is independent of what other types have positive probability ) . provides a generalization of the above result for the two long - run player case in a special class of games called of `` conflicting interests '' , when one of the players is sufficiently more patient than the opponent .
a game is of conflicting interests when the commitment strategy of one player ( @xmath0 ) holds the opponent ( @xmath2 ) to his minimax payoff .
the author derives an upper limit on the number of rounds @xmath2 will not play a best response to @xmath0 s commitment type , which in turn generates a lower bound on @xmath0 s equilibrium payoff . for a detailed treatment of the reputation effect , the reader is directed to the work of . in computer science and information systems research ,
_ reputation _ information defines some aggregate of feedback reports about past transactions .
this is the semantics we are using when referring to the reputation of the provider .
reputation information encompasses a unitary appreciation of the personal attributes of the provider , and influences the trusting decisions of clients . depending on the environment , reputation has two main roles : to _ signal _ the capabilities of the provider , and to _ sanction _ cheating behavior @xcite .
_ signaling _ reputation mechanisms allow clients to learn which providers are the most capable of providing good service .
such systems have been widely used in computational trust mechanisms . and describe systems where agents use their direct past experience to recognize trustworthy partners .
the global efficiency of the market is clearly increased , however , the time needed to build the reputation information prohibits the use of this kind of mechanisms in a large scale online market . a number of signaling reputation mechanisms also take into consideration indirect reputation information , i.e. , information reported by peers .
and use social networks in order to obtain the reputation of an unknown agent .
agents ask acquaintances several hops away about the trustworthiness of an unknown agent .
recommendations are afterwards aggregated into a single measure of the agent s reputation .
this class of mechanisms , however intuitive , does not provide any rational participation incentives for the agents .
moreover , there is little protection against untruthful reporting , and no guarantee that the mechanism can not be manipulated by a malicious provider in order to obtain higher payoffs .
truthful reporting incentives for signaling reputation mechanisms are described by .
honest reports are explicitly rewarded by payments that take into account the value of the submitted report , and the value of a report submitted by another client ( called the _ reference reporter _ ) .
the payment schemes are designed based on _ proper scoring rules _ , mathematical functions that make possible the revelation of private beliefs @xcite . the essence behind honest reporting incentives is the observation that the private information a client obtains from interacting with a provider changes her belief regarding the reports of other clients . this change in beliefs can be exploited to make honesty an ex - ante nash equilibrium strategy .
extend the above result by taking a computational approach to designing incentive compatible payment schemes . instead of using closed form scoring rules
, they compute the payments using an optimization problem that minimizes the total budget required to reward the reporters . by also using several reference reports and filtering mechanisms
, they render the payment mechanisms cheaper and more practical .
presents a comprehensive investigation of binary _ sanctioning _ reputation mechanisms . as in our setting , providers are equally capable of providing high quality , however , doing so requires costly effort .
the role of the reputation mechanism is to encourage cooperative behavior by punishing cheating : negative feedback reduces future revenues either by excluding the provider from the market , or by decreasing the price the provider can charge in future transactions .
dellarocas shows that simple information structures and decision rules can lead to efficient equilibria , given that clients report honestly .
our paper builds upon such mechanisms by addressing reporting incentives .
we will abstract away the details of the underlying reputation mechanism through an explicit penalty associated with a negative feedback . given that such high enough penalties exist ,
any reputation mechanism ( i.e. , feedback aggregation and trusting decision rules ) can be plugged in our scheme . in the same group of work that addresses reporting incentives
, we mention the work of , and .
@xcite consider exchanges of goods for money and prove that a market in which agents are trusted to the degree they deserve to be trusted is equally efficient as a market with complete trustworthiness . by scaling the amount of the traded product ,
the authors prove that it is possible to make it rational for sellers to truthfully declare their trustworthiness .
truthful declaration of one s trustworthiness eliminates the need of reputation mechanisms and significantly reduces the cost of trust management .
however , the assumptions made about the trading environment ( i.e. the form of the cost function and the selling price which is supposed to be smaller than the marginal cost ) are not common in most electronic markets . for e - bay - like auctions , the goodwill hunting mechanism @xcite provides a way to make sellers indifferent between lying or truthfully declaring the quality of the good offered for sale . momentary gains or losses obtained from misrepresenting the good s quality
are later compensated by the mechanism which has the power to modify the announcement of the seller .
describe an incentive - compatible reputation mechanism that is particularly suited for peer - to - peer applications .
their mechanism is similar to ours , in the sense that both the provider and the client are punished for submitting conflicting reports .
the authors experimentally show that a class of common lying strategies are successfully deterred by their scheme . unlike their results , our paper considers _ all _ possible equilibrium strategies and sets bounds on the amount of untruthful information recorded by the reputation mechanism .
we assume an online market , where rational clients ( she ) repeatedly request the same service from one provider ( he ) .
every client repeatedly interacts with the service provider , however , successive requests from the same client are always interleaved with enough requests generated by other clients .
transactions are assumed sequential , the provider does not have capacity constraints , and accepts all requests .
the price of service is @xmath3 monetary units , and the service can have either high ( @xmath4 ) or low ( @xmath5 ) quality .
only high quality is valuable to the clients , and has utility @xmath6 .
low quality has utility 0 , and can be precisely distinguished from high quality . before each round , the client can decide to request the service from the provider , or quit the market and resort to an outside provider that is completely trustworthy .
the outside provider always delivers high quality service , but for a higher price @xmath7 .
if the client decides to interact with the online provider , she issues a request to the provider , and pays for the service . the provider can now decide to exert low ( @xmath8 ) or high ( @xmath9 ) effort when treating the request .
low effort has a normalized cost of 0 , but generates only low quality .
high effort is expensive ( normalized cost equals @xmath10 ) and generates high quality with probability @xmath11 .
@xmath12 is fixed , and depends on the environmental factors outside the control of the provider .
@xmath13 , so that it is individually rational for providers to exert effort . after exerting effort
, the provider can observe the quality of the resulting service .
he can then decide to deliver the service as it is , or to acknowledge failure and roll back the transaction by fully reimbursing the client .
we assume perfect delivery channels , such that the client perceives exactly the same quality as the provider .
after delivery , the client inspects the quality of service , and can accuse low quality by submitting a negative report to the reputation mechanism . the reputation mechanism ( rm )
is unique in the market , and trusted by all participants .
it can oversee monetary transactions ( i.e. , payments made between clients and the provider ) and can impose fines on all parties .
however , the rm does not observe the effort level exerted by the provider , nor does it know the quality of the delivered service
. the rm asks feedback from the client only if she chose to transact with the provider in the current round ( i.e. , paid the price of service to the provider ) and the provider delivered the service ( i.e. , provider did not reimburse the client ) .
when the client submits negative feedback , the rm punishes both the client and the provider : the client must pay a fine @xmath14 , and the provider accumulates a negative reputation report . although simplistic , this model retains the main characteristics of several interesting applications .
a delivery service for perishable goods ( goods that lose value past a certain deadline ) is one of them .
pizza , for example , must be delivered within 30 minutes , otherwise it gets cold and loses its taste .
hungry clients can order at home , or drive to a more expensive local restaurant , where they re sure to get a hot pizza .
the price of a home delivered pizza is @xmath15 , while at the restaurant , the same pizza would cost @xmath16 . in both cases ,
the utility of a warm meal is @xmath17 .
the pizza delivery provider must exert costly effort to deliver orders within the deadline .
a courier must be dispatched immediately ( high effort ) , for an estimated cost of @xmath18 .
while such action usually results in good service ( the probability of a timely delivery is @xmath19 ) , traffic conditions and unexpected accidents ( e.g. , the address is not easily found ) may still delay some deliveries past the deadline .
once at the destination , the delivery person , as well as the client , know if the delivery was late or not . as it is common practice , the provider can acknowledge being late , and reimburse the client .
clients may provide feedback to a reputation mechanism , but their feedback counts only if they were not reimbursed .
the client s fine for submitting a negative report can be set for example at @xmath20 .
the future loss to the provider caused by the negative report ( and quantified through @xmath21 ) depends on the reputation mechanism .
a simplified market of car garagists or plumbers could fit the same model .
the provider is commissioned to repair a car ( respectively the plumbing ) and the quality of the work depends on the exerted effort .
high effort is more costly but ensures a lasting result with high probability .
low effort is cheap , but the resulting fix is only temporary . in both cases , however , the warranty convention may specify the right of the client to ask for a reimbursement if problems reoccur within the warranty period
. reputation feedback may be submitted at the end of the warranty period , and is accepted only if reimbursements did nt occur .
an interesting emerging application comes with a new generation of web services that can optimally decide how to treat every request . for some service types ,
a high quality response requires the exclusive use of costly resources .
for example , computation jobs require cpu time , storage requests need disk space , information requests need queries to databases .
sufficient resources , is a prerequisite , but not a guarantee for good service .
software and hardware failures may occur , however , these failures are properly signaled to the provider .
once monetary incentives become sufficiently important in such markets , intelligent providers will identify the moral hazard problem , and may act strategically as identified in our model .
from game theoretic point of view , one interaction between the client and the provider can be modeled by the extensive - form game ( @xmath22 ) with imperfect public information , shown in figure [ fig : game ] .
the client moves first and decides ( at node 1 ) whether to play @xmath23 and interact with the provider , or to play @xmath24 and resort to the trusted outside option .
once the client plays @xmath23 , the provider can chose at node 2 whether to exert high or low effort ( i.e. , plays @xmath9 or @xmath8 respectively ) .
when the provider plays @xmath8 the generated quality is low .
when the provider plays @xmath9 , nature chooses between high quality ( @xmath4 ) with probability @xmath12 , and low quality ( @xmath5 ) with probability @xmath25 .
the constant @xmath12 is assumed common knowledge in the market .
having seen the resulting quality , the provider delivers ( i.e. , plays @xmath26 ) the service , or acknowledges low quality and rolls back the transaction ( i.e. , plays @xmath27 ) by fully reimbursing the client . if the service is delivered , the client can report positive ( @xmath28 ) or negative ( @xmath29 ) feedback .
a pure strategy is a deterministic mapping describing an action for each of the player s information sets .
the client has three information sets in the game @xmath22 .
the first information set is singleton and contains the node 1 at the beginning of game when the client must decide between playing @xmath23 or @xmath24 .
the second information set contains the nodes 7 and 8 ( the dotted oval in figure @xmath30 ) where the client must decide between reporting @xmath29 or @xmath28 , given that she has received low quality , @xmath5 .
the third information set is singleton and contains the node 9 where the client must decide between reporting @xmath29 or @xmath28 , given that she received high quality , @xmath4 .
the strategy @xmath31 , for example , is the honest reporting strategy , specifying that the client enters the game , reports @xmath29 when she receives low quality , and reports @xmath28 when she receives high quality .
the set of pure strategies of the client is : @xmath32 similarly , the set of pure strategies of the provider is : @xmath33 where @xmath34 , for example , is the socially desired strategy : the provider exerts effort at node 2 , acknowledges low quality at node 5 , and delivers high quality at node 6 .
a pure strategy profile @xmath35 is a pair @xmath36 where @xmath37 and @xmath38 .
if @xmath39 denotes the set of probability distributions over the elements of @xmath0 , @xmath40 and @xmath41 are mixed strategies for the client , respectively the provider , and @xmath42 is a mixed strategy profile .
the payoffs to the players depend on the chosen strategy profile , and on the move of nature .
let @xmath43 denote the pair of expected payoffs received by the client , respectively by the provider when playing strategy profile @xmath1 .
the function @xmath44 is characterized in table [ tab : normalform ] and also describs the normal form transformation of @xmath22 . besides the corresponding payments made between the client and the provider , table [ tab : normalform ] also reflects the influence of the reputation mechanism , as further explained in section [ reputation_mech ] .
the four strategies of the client that involve playing @xmath24 at node 1 generate the same outcomes , and therefore , have been collapsed for simplicity into a single row of table [ tab : normalform ] .
cc provider & .normal transformation of the extensive form game , @xmath22 [ cols="^,^,^,^,^,^ " , ] for every interaction , the reputation mechanism records one of the three different signals it may receive : _ positive _ feedback when the client reports @xmath28 , _ negative _ feedback when the client reports @xmath29 , and _
neutral _ feedback when the provider rolls back the transaction and reimburses the client . in figure
[ fig : game ] ( and table [ tab : normalform ] ) positive and neutral feedback do not influence the payoff of the provider , while negative feedback imposes a punishment equivalent to @xmath21 .
two considerations made us choose this representation .
first , we associate neutral and positive feedback with the same reward ( 0 in this case ) because intuitively , the acknowledgement of failure may also be regarded as `` honest '' behavior on behalf of the provider .
failures occur despite best effort , and by acknowledging them , the provider should nt suffer .
however , neutral feedback may also result because the provider did not exert effort .
the lack of punishment for these instances contradicts the goal of the reputation mechanism to encourage exertion of effort .
fortunately , the action @xmath45 can be the result of rational behavior only in two circumstances , both excusable : one , when the provider defends himself against a malicious client that is expected to falsely report negative feedback ( details in section [ evilbuyers ] ) , and two , when the environmental noise is too big ( @xmath12 is too small ) to justify exertion of effort .
neutral feedback can be used to estimate the parameter @xmath12 , or to detect coalitions of malicious clients , and indirectly , may influence the revenue of the provider . however , for the simplified model presented above , positive and neutral feedback are considered the same in terms of generated payoffs .
the second argument relates to the role of the rm to constrain the revenue of the provider depending on the feedback of the client .
there are several ways of doing that
. describes two principles , and two mechanisms that punish the provider when the clients submit negative reports .
the first , works by exclusion . after each negative report
the reputation mechanism bans the provider from the market with probability @xmath46 .
this probability can be tuned such that the provider has the incentive to cooperate almost all the time , and the market stays efficient .
the second works by changing the conditions of future trade .
every negative report triggers the decrease of the price the next @xmath47 clients will pay for the service . for lower values of @xmath47
the price decrease is higher , nonetheless , @xmath47 can take any value in an efficient market .
both mechanisms work because the future losses offset the momentary gain the provider would have had by intentionally cheating on the client .
note that these penalties are given endogenously by lost future opportunities , and require some minimum premiums for trusted providers .
when margins are not high enough , providers do not care enough about future transactions , and will use the present opportunity of cheating .
another option is to use exogenous penalties for cheating .
for example , the provider may be required to buy a licence for operating in the market .
the licence is _ partially destroyed _ by every negative feedback .
totaly destroyed licences must be restored through a new payment , and remaining parts can be sold if the provider quits the market . the price of the licence and the amount that is destroyed by a negative feedback can be scaled such that rational providers have the incentive to cooperate . unlike the previous solutions , this mechanism does not require minimum transaction margins as punishments for negative feedback are directly subtracted from the upfront deposit . one way or another , all reputation mechanisms foster cooperation because the provider associates value to client feedback .
let @xmath48 and @xmath49 be the value of a positive , respectively a negative report . in the game in figure
[ fig : game ] , @xmath48 is normalized to 0 , and @xmath49 is @xmath21 . by using this notation , we abstract away the details of the reputation mechanism , and retain only the essential punishment associated with negative feedback .
any reputation mechanism can be plugged in our scheme , as long as the particular constraints ( e.g. , minimum margins for transactions ) are satisfied .
one last aspect to be considered is the influence of the reputation mechanism on the future transactions of the client .
if negative reports attract lower prices , rational long - run clients might be tempted to falsely report in order to purchase cheaper services in the future .
fortunately , some of the mechanisms designed for single - run clients , do not influence the reporting strategy of long - run clients .
the reputation mechanism that only keeps the last @xmath47 reports @xcite is one of them .
a false negative report only influences the next @xmath47 transactions of the provider ; given that more than @xmath47 other requests are interleaved between any two successive requests of the same client , a dishonest reporter can not decrease the price for her future transactions .
the licence - based mechanism we have described above is another example .
the price of service remains unchanged , therefore reporting incentives are unaffected . on the other hand , when negative feedback is punished by exclusion , clients may be more reluctant to report negatively , since they also lose a trading partner .
the one - time game presented in figure [ fig : game ] has only one subgame equilibrium where the client opts @xmath24 . when asked to report feedback
, the client always prefers to report @xmath28 ( reporting @xmath29 attracts the penalty @xmath14 ) . knowing this , the best strategy for the provider is to exert low effort and deliver the service . knowing the provider will play @xmath50 , it is strictly better for the client to play @xmath24 .
the repeated game between the same client and provider may , however , have other equilibria . before analyzing the repeated game ,
let us note that every interaction between a provider and a particular client can be strategically isolated and considered independently .
as the provider accepts all clients and views them identically , he will maximize his expected revenue in each of the isolated repeated games . from now on
, we will only consider the repeated interaction between the provider and one client .
this can be modeled by a repetition of the stage game @xmath22 , denoted @xmath51 , where @xmath52 is finite or infinite . in this paper
we will deal with the infinite horizon case , however , the results obtained can also be applied with minor modifications to finitely repeated games where @xmath52 is large enough .
if @xmath53 is the per period discount factor reflecting the probability that the market ceases to exist after each round , ( or the present value of future revenues ) , let us denote by @xmath54 the expected discount factor in the game @xmath51 . if our client interacts with the provider on the average every @xmath47 rounds , @xmath55
. the life - time expected payoff of the players is computed as : @xmath56 where @xmath57 is the client , respectively the provider , @xmath58 is the expected payoff obtained by player @xmath59 in the @xmath60 interaction , and @xmath61 is the discount applied to compute the present day value of @xmath58
. we will consider _ normalized _ life - time expected payoffs , so that payoffs in @xmath22 and @xmath51 can be expressed using the same measure : @xmath62 we define the _ average continuation payoff _ for player @xmath59 from period @xmath63 onward ( and including period @xmath63 ) as : @xmath64 the set of outcomes publicly perceived by both players after each round is : @xmath65 where : * @xmath24 is observed when the client opts @xmath24 , * @xmath27 is observed when the provider acknowledges low quality and rolls back the transaction , * @xmath66 is observed when the provider delivers quality @xmath67 and the client reports @xmath68 .
we denote by @xmath69 a specific public history of the repeated game out of the set @xmath70 of all possible histories up to and including period @xmath63 . in the repeated game ,
a public strategy @xmath71 of player @xmath59 is a sequence of maps @xmath72 , where @xmath73 prescribes the ( mixed ) strategy to be played in round @xmath63 , after the public history @xmath74
. a _ perfect public equilibrium _
( ppe ) is a profile of public strategies @xmath75 that , beginning at any time @xmath63 and given any public history @xmath76 , form a nash equilibrium from that point on @xcite .
@xmath77 is the continuation payoff to player @xmath59 given by the strategy profile @xmath1 .
@xmath22 is a game with _ product structure _ since any public outcome can be expressed as a vector of two components @xmath78 such that the distribution of @xmath79 depends only on the actions of player @xmath57 , the client , respectively the provider . for such games ,
establish a folk theorem proving that any feasible , individually rational payoff profile is achievable as a ppe of @xmath80 when the discount factor is close enough to 1 .
the set of feasible , individually rational payoff profiles is characterized by : * the minimax payoff to the client , obtained by the option @xmath24 : @xmath81 ; * the minimax payoff to the provider , obtained when the provider plays @xmath45 : @xmath82 ; * the pareto optimal frontier ( graphically presented in figure [ fig : paretooptfront ] ) delimited by the payoffs given by ( linear combination of ) the strategy profiles @xmath83 , @xmath84 , @xmath85 and @xmath86 . and contains more than one point ( i.e. , the payoff when the client plays @xmath24 ) when @xmath87 and @xmath88 .
both conditions impose restrictions on the minimum margin generated by a transaction such that the interaction is profitable .
the ppe payoff profile that gives the provider the maximum payoff is @xmath89 where : @xmath90 and @xmath91 is defined above . .
] while completely characterizing the set of ppe payoffs for discount factors strictly smaller than 1 is outside the scope of this paper , let us note the following results : first , if the discount factor is high enough ( but strictly less than 1 ) with respect to the profit margin obtained by the provider from one interaction , there is at least one ppe such that the reputation mechanism records only honest reports .
moreover , this equilibrium is pareto - optimal . when @xmath92 , the strategy profile : * the provider always exerts high effort , and delivers only high quality ; if the client deviates from the equilibrium , the provider switches to @xmath50 for the rest of the rounds ; * the client always reports @xmath28 when asked to submit feedback ; if the provider deviates , ( i.e. , she receives low quality ) , the client switches to @xmath24 for the rest of the rounds . is a pareto - optimal ppe .
[ prop : lowerbounddelta ] it is not profitable for the client to deviate from the equilibrium path .
reporting @xmath29 attracts the penalty @xmath14 in the present round , and the termination of the interaction with the provider ( the provider stops exerting effort from that round onwards ) .
the provider , on the other hand , can momentarily gain by deviating to @xmath93 or @xmath50 .
a deviation to @xmath93 gives an expected momentary gain of @xmath94 and an expected continuation loss of @xmath95 .
a deviation to @xmath50 brings an expected momentary gain equal to @xmath96 and an expected continuation loss of @xmath97 . for the discount factor satisfying our hypothesis ,
both deviations are not profitable .
the discount factor is low enough with respect to profit margins , such that the future revenues given by the equilibrium strategy offset the momentary gains obtained by deviating .
the equilibrium payoff profile is @xmath98 , which is pareto - optimal and socially efficient .
second , we can prove that the client never reports negative feedback in any pareto - optimal ppe , regardless the value of the discount factor .
the restriction to pareto - optimal is justifiable by practical reasons : assuming that the client and the provider can somehow negotiate the equilibrium they are going to play , it makes most sense to choose one of the pareto - optimal equilibria .
the probability that the client reports negative feedback on the equilibrium path of any pareto - optimal ppe strategy is zero .
[ prop : nozero ] the full proof presented in appendix [ ap : nozero ] follows the following steps .
step 1 , all equilibrium payoffs can be expressed by adding the present round payoff to the discounted continuation payoff from the next round onward .
step 2 , take the ppe payoff profile @xmath99 , such that there is no other ppe payoff profile @xmath100 with @xmath101 .
the client never reports negative feedback in the first round of the equilibrium that gives @xmath102 .
step 3 , the equilibrium continuation payoff after the first round also satisfies the conditions set for @xmath102 .
hence , the probability that the client reports negative feedback on the equilibrium path that gives @xmath102 is 0 .
pareto - optimal ppe payoff profiles clearly satisfy the definition of @xmath102 , hence the result of the proposition .
the third result we want to mention here , is that there is an upper bound on the percentage of false reports recorded by the reputation mechanism in any of the pareto - optimal equilibria .
the upper bound on the percentage of false reports recorded by the reputation mechanism in any ppe equilibrium is : @xmath103 [ prop : boundnorep ] the full proof presented in appendix [ ap : boundnorep ] builds directly on the result of proposition [ prop : nozero ] . since clients never report negative feedback along pareto - optimal equilibria , the only false reports recorded by the reputation mechanism appear when the provider delivers low quality , and the client reports positive feedback
. however , any ppe profile must give the client at least @xmath81 , otherwise the client is better off by resorting to the outside option .
every round in which the provider deliberatively delivers low quality gives the client a payoff strictly smaller than @xmath104 .
an equilibrium payoff greater than @xmath91 is therefore possible only when the percentage of rounds where the provider delivers low quality is bounded .
the same bound limits the percentage of false reports recorded by the reputation mechanism . for a more intuitive understanding of the results presented in this section ,
let us refer to the pizza delivery example detailed in section [ setting_example ] .
the price of a home delivered pizza is @xmath15 , while at the local restaurant the same pizza would cost @xmath16 .
the utility of a warm pizza to the client is @xmath105 , the cost of delivery is @xmath18 and the probability that unexpected traffic conditions delay the delivery beyond the 30 minutes deadline ( despite the best effort of the provider ) is @xmath106 .
the client can secure a minimax payoff of @xmath107 by always going out to the restaurant .
however , the socially desired equilibrium happens when the client orders pizza at home , and the pizza service exerts effort to deliver pizza in time : in this case the payoff of the client is @xmath108 , while the payoff of the provider is @xmath109 .
proposition [ prop : lowerbounddelta ] gives a lower bound on the discount factor of the pizza delivery service such that repeated clients can expect the socially desired equilibrium .
this bound is @xmath110 ; assuming that the daily discount factor of the pizza service is @xmath111 , the same client must order pizza at home at least once every 6 weeks .
the values of the discount factors can also be interpreted in terms of the minimum number of rounds the client ( and the provider ) will likely play the game .
for example , the discount factor can be viewed as the probability that the client ( respectively the provider ) will `` live '' for another interaction in the market .
it follows that the average lifetime of the provider is at least @xmath112 interactions ( with all clients ) , while the average lifetime of the client is at least @xmath113 interactions ( with the same pizza delivery service ) .
these are clearly realistic numbers .
proposition [ prop : boundnorep ] gives an upper bound on the percentage of false reports that our mechanism may record in equilibrium from the clients .
as @xmath114 , this limit is : @xmath115 it follows that at least @xmath116 of the reports recorded by our mechanism ( in any equilibrium ) are correct .
the false reports ( false positive reports ) result from rare cases where the pizza delivery is intentionally delayed to save some cost but clients do not complain .
the false report can be justified , for example , by the provider s threat to refuse future orders from clients that complain . given that late deliveries are still rare enough , clients are better off with the home delivery than with the restaurant , hence they accept the threat . as other options become available to the clients ( e.g. , competing delivery services ) the bound @xmath117 will decrease . please note that the upper bound defined by proposition [ prop : boundnorep ] only depends on the outside alternative available to the provider , and is not influenced by the punishment @xmath21 introduced by the reputation mechanism .
this happens because the revenue of a client is independent of the interactions of other clients , and therefore , on the reputation information as reported by other clients .
equilibrium strategies are exclusively based on the direct experience of the client . in the following section , however
, we will refine this bound by considering that clients can build a reputation for reporting honestly .
there , the punishment @xmath21 plays an important role .
an immediate consequence of propositions [ prop : nozero ] and [ prop : boundnorep ] is that the provider can extract all of the surplus created by the transactions by occasionally delivering low quality , and convincing the clients not to report negative feedback ( providers can do so by promising sufficiently high continuation payoffs that prevent the client to resort to the outside provider ) . assuming that the provider has more `` power '' in the market , he could influence the choice of the equilibrium strategy to one that gives him the most revenue , and holds the clients close to the minimax payoff @xmath81 given by the outside option . :
the continuation payoffs enforcing a pareto - optimal ppe payoff profile are also pareto - optimal .
therefore , clients falsely report positive feedback even under the more restrictive notion of negotiation - proof equilibrium . ] however , a client who could commit to report honestly , ( i.e. , commit to play the strategy @xmath118 ) would benefit from cooperative trade .
the provider s best response against @xmath119 is to play @xmath120 repeatedly , which leads the game to the socially efficient outcome .
unfortunately the commitment to @xmath121 is not credible in the complete information game , for the reasons explained in section [ eqanalysis ] . following the results of , and we know that such honest reporting commitments may become credible in a game with incomplete information .
suppose that the provider has incomplete information in @xmath122 , and believes with non - negative probability that he is facing a committed client that always reports the truth . a rational client
can then `` fake '' the committed client , and `` build a reputation '' for reporting honestly . when the reputation becomes credible
, the provider will play @xmath34 ( the best response against @xmath121 ) , which is better for the client than the payoff she would obtain if the provider knew she was the `` rational '' type . as an effect of reputation building
, the set of equilibrium points is reduced to a set where the payoff to the client is higher than the payoff obtained by a client committed to report honestly .
as anticipated from proposition [ prop : boundnorep ] , a smaller set of equilibrium points also reduces the bound of false reports recorded by the reputation mechanism . in certain cases , this bound can be reduced to almost zero .
formally , incomplete information can be modeled by a perturbation of the complete information repeated game @xmath122 such that in period 0 ( before the first round of the game is played ) the `` type '' of the client is drawn by nature out of a countable set @xmath123 according to the probability measure @xmath124 .
the client s payoff now additionally depends on her type .
we say that in the perturbed game @xmath125 the provider has incomplete information because he is not sure about the true type of the client .
two types from @xmath123 have particular importance : * the `` normal '' type of the client , denoted by @xmath126 , is the rational client who has the payoffs presented in figure [ fig : game ] . * the `` commitment '' type of the client , denoted by @xmath127 , always prefers to play the commitment strategy @xmath121 . from a rational perspective , the commitment type client obtains an arbitrarily high supplementary reward for reporting the truth .
this external reward makes the strategy @xmath121 the dominant strategy , and therefore , no commitment type client will play anything else than @xmath121 . in theorem [
th : equilibrium ] we give an upper bound @xmath128 on the number of times the provider delivers low quality in @xmath125 , given that he always observes the client reporting honestly .
the intuition behind this result is the following .
the provider s best response to a honest reporter is @xmath34 : always exert high effort , and deliver only when the quality is high .
this gives the commitment type client her maximum attainable payoff in @xmath125 , corresponding to the socially efficient outcome . the provider , however , would be better off by playing against the normal type client , against whom he can obtain an expected payoff greater than @xmath97 . the normal type client may be distinguished from a commitment type client only in the rounds when the provider delivers low quality : the commitment type always reports negative feedback , while the normal type might decide to report positive feedback in order to avoid the penalty @xmath14 .
the provider can therefore decide to deliver low quality to the client in order to test her real type .
the question is , how many times should the provider test the true type of the client . every failed test ( i.e. , the provider delivers low quality and the client reports negative feedback ) generates a loss of @xmath129 to the provider , and slightly enforces the belief that the client reports honestly .
since the provider can not wait infinitely for future payoffs , there must be a time when the provider will stop testing the type of the provider , and accepts to play the socially efficient strategy , @xmath34 .
the switch to the socially efficient strategy is not triggered by a revelation of the client s type .
the provider believes that the client _ behaves _ as if she were a commitment type , not that the client _ is _ a commitment type .
the client may very well be a normal type who chooses to mimic the commitment type , in the hope that she will obtain better service from the provider .
however , further trying to determine the true type of the client is too costly for the provider . therefore , the provider chooses to play @xmath120 , which is the best response to the commitment strategy @xmath121 .
if the provider has incomplete information in @xmath122 , and assigns positive probability to the normal and commitment type of the client ( @xmath130 , @xmath131 ) , there is a finite upper bound , @xmath128 , on the number of times the provider delivers low quality in any equilibrium of @xmath125 .
this upper bound is : @xmath132 [ th : equilibrium ] first , we use an important result obtained by about statistical inference ( lemma 1 ) : if every previously delivered low quality service was sanctioned by a negative report , the provider must expect with increasing probability that his next low quality delivery will also be sanctioned by negative feedback .
technically , for any @xmath133 , the provider can deliver at most @xmath134 low quality services ( sanctioned by negative feedback ) before expecting that the @xmath135 low quality delivery will also be sanctioned by negative feedback with probability greater then @xmath46 .
this number equals to : @xmath136 as stated earlier , this lemma does not prove that the provider will become convinced that he is facing a commitment type client .
it simply proves that after a finite number of rounds the provider becomes convinced that the client is playing as if she were a commitment type .
second , if @xmath137 but is strictly smaller than @xmath28 , the rational provider does not deliver low quality ( it is easy to verify that the maximum discounted future gain does not compensate for the risk of getting a negative feedback in the present round ) . by the previously mentioned lemma , it must be that in any equilibrium , the provider delivers low quality a finite number of times .
third , let us analyze the round , @xmath138 , when the provider is about to deliver a low quality service ( play @xmath139 ) for the last time .
if @xmath46 is the belief of the provider that the client reports honestly in round @xmath138 , his expected payoff ( just before deciding to deliver the low quality service ) can be computed as follows : * with probability @xmath46 the client reports 0 .
her reputation for reporting honestly becomes credible , so the provider plays @xmath34 in all subsequent rounds .
the provider gains @xmath140 in the current round , and expects @xmath141 for the subsequent rounds ; * with probability @xmath142 , the client reports @xmath28 and deviates from the commitment strategy , the provider knows he is facing a rational client , and can choose a continuation ppe strategy from the complete information game .
he gains @xmath3 in the current round , and expects at most @xmath143 in the subsequent rounds ; @xmath144 on the other hand , had the provider acknowledged the low quality and rolled back the transaction ( i.e. , play @xmath145 ) , his expected payoff would have been at least : @xmath146 since the provider chooses nonetheless to play @xmath139 it must be that @xmath147 which is equivalent to : @xmath148 finally , by replacing equation ( [ eq : overlinepi ] ) in the definition of @xmath134 we obtain the upper bound on the number of times the provider delivers low quality service to a client committed to report honestly .
the existence of @xmath128 further reduces the possible equilibrium payoffs a client can get in @xmath149 .
consider a rational client who receives for the first time low quality .
she has the following options : * report negative feedback and attempt to build a reputation for reporting honestly .
her payoff for the current round is @xmath150 .
moreover , her worst case expectation for the future is that the next @xmath151 rounds will also give her @xmath150 , followed by the commitment payoff equal to @xmath152 : + @xmath153 * on the other hand , by reporting positive feedback she reveals to be a normal type , loses only @xmath3 in the current round , and expects a continuation payoff equal to @xmath154 given by a ppe strategy profile of the complete information game @xmath80 : + @xmath155 the reputation mechanism records false reports only when clients do not have the incentive to build a reputation for reporting honestly , and @xmath156 ; this is true for : @xmath157 following the argument of proposition [ prop : boundnorep ] we can obtain a bound on the percentage of false reports recorded by the reputation mechanism in a pareto - optimal ppe that gives the client at least @xmath154 : @xmath158 of particular importance is the case when @xmath159 .
@xmath154 and @xmath160 become : @xmath161 so the probability of recording a false report ( after the first one ) can be arbitrarily close to 0 as @xmath162 . for the pizza delivery example introduced in section [ setting_example ] , figure [ fig : k_p ] plots
the bound , @xmath128 , defined in theorem [ th : equilibrium ] , as a function of the prior belief ( @xmath163 ) of the provider that the client is an honest reporter . we have used a value of the discount factor equal to @xmath164 , such that on average , every client interacts @xmath165 times with the same provider . the penalty for negative feedback
was taken @xmath166 . when the provider believes that @xmath167 of the clients always report honestly
, he will deliver at most 3 times low quality .
when the belief goes up to @xmath168 no rational provider will deliver low quality more than once . as a function of the prior belief @xmath169 . ] in figure [ fig : gamma ] we plot the values of the bounds @xmath117 ( equation ( [ eq : boundnorep ] ) ) and @xmath160 ( equation ( [ eq : boundwithrep ] ) ) as a function of the prior belief @xmath169 .
the bounds simultaneously hold , therefore the maximum percentage of false reports recorded by the reputation mechanism is the minimum of the two .
when @xmath169 is less @xmath170 , @xmath171 , @xmath172 , and the reputation effect does not significantly reduce the worst case percentage of false reports recorded by the mechanism . however , when @xmath173 the reputation mechanism records ( in the worst case ) only half as many false reports , and as @xmath174 , the percentage of false reports drops to @xmath175 .
this probability can be further decreased by decreasing the penalty @xmath14 . in the limit , as @xmath14 approaches 0
, the reputation mechanism will register a false report with vanishing probability .
the result of theorem [ th : equilibrium ] has to be interpreted as a worst case scenario . in real markets , providers that already have a small predisposition to cooperate
will defect fewer times .
moreover , the mechanism is self enforcing , in the sense that the more clients act as commitment types , the higher will be the prior beliefs of the providers that new , unknown clients will report truthfully , and therefore the easier it will be for the new clients to act as truthful reporters .
as mentioned at the end of section [ eqanalysis ] , the bound @xmath160 strongly depends on the punishment @xmath21 imposed by the reputation mechanism for a negative feedback .
the higher @xmath21 , the easier it is for clients to build a reputation , and therefore , the lower the amount of false information recorded by the reputation mechanism . .
the mechanism described so far encourages service providers to do their best and deliver good service .
the clients were assumed rational , or committed to report honestly , and in either case , they never report negative feedback unfairly . in this section
, we investigate what happens when clients explicitly try to `` hurt '' the providers by submitting fake negative ratings to the reputation mechanism .
an immediate consequence of fake negative reports is that clients lose money .
however , the costs @xmath14 of a negative report would probably be too small to deter clients with separate agendas from hurting the provider .
fortunately , the mechanism we propose naturally protects service providers from consistent attacks initiated by malicious clients .
formally , a _
malicious type _
client , @xmath176 , obtains a supplementary ( external ) payoff @xmath177 for reporting negative feedback .
obviously , @xmath177 has to be greater than the penalty @xmath14 , otherwise the results of proposition [ prop : nozero ] would apply . in the incomplete information game @xmath149 , the provider now assigns non - zero initial probability to the belief that the client is malicious .
when only the normal type , @xmath126 , the honest reporter type @xmath127 and the malicious type @xmath178 have non - zero initial probability , the mechanism we describe is robust against unfair negative reports .
the first false negative report exposes the client as being malicious , since neither the normal , nor the commitment type report @xmath29 after receiving high quality . by bayes law ,
the provider s updated belief following a false negative report must assign probability 1 to the malicious type . although providers are not allowed to refuse service requests , they can protect themselves against malicious clients by playing @xmath45 : i.e. , exert low effort and reimburse the client afterwards .
the rm records neutral feedback in this case , and does not sanction the provider . against @xmath45 ,
malicious clients are better off by quitting the market ( opt @xmath24 ) , thus stopping the attack .
the rm records at most one false negative report for every malicious client , and assuming that identity changes are difficult , providers are not vulnerable to unfair punishments . when other types ( besides @xmath179 and @xmath178 ) have non - zero initial probability , malicious clients are harder to detect .
they could masquerade client types that are normal , but accidentally misreport .
it is not rational for the provider to immediately exclude ( by playing @xmath45 ) normal clients that rarely misreport : the majority of the cooperative transactions rewarded by positive feedback still generate positive payoffs .
let us now consider the client type @xmath180 that behaves exactly like the normal type , but misreports @xmath29 instead of @xmath28 independently with probability @xmath181 . when interacting with the client type @xmath182 , the provider receives the maximum number of unfair negative reports when playing the efficient equilibrium : i.e. , @xmath34 . in this case
, the provider s expected payoff is : @xmath183 since @xmath184 has to be positive ( the minimax payoff of the provider is 0 , given by @xmath45 ) , it must be that @xmath185 . the maximum value of @xmath181 is also a good approximation for the maximum percentage of false negative reports the malicious type can submit to the reputation mechanism .
any significantly higher number of harmful reports exposes the malicious type and allows the provider to defend himself .
note , however , that the malicious type can submit a fraction @xmath181 of false reports only when the type @xmath182 has positive prior probability .
when the provider does not believe that a normal client can make so many mistakes ( even if the percentage of false reports is still low enough to generate positive revenues ) he attributes the false reports to a malicious type , and disengages from cooperative behavior .
therefore , one method to reduce the impact of malicious clients is to make sure that normal clients make few or no mistakes .
technical means ( for example by providing automated tools for formatting and submitting feedback ) , or improved user interfaces ( that make it easier for human users to spot reporting mistakes ) will greatly limit the percentage of mistakes made by normal clients , and therefore , also reduce the amount of harm done by malicious clients .
one concrete method for reducing mistakes is to solicit only negative feedback from the clients ( the principle that no news is good news , also applied by ) .
as reporting involves some conscious decision , mistakes will be less frequent .
on the other hand , the reporting effort will add to the penalty for a negative report , and makes it harder for normal clients to establish a reputation for honest reporters .
alternative methods for reducing the harm done by malicious clients ( like filtering mechanisms , etc . , ) as well as tighter bounds on the percentage of false reports introduced by such clients will be further addressed in future work .
further benefits can be obtained if the clients reputation for reporting honestly is shared within the market .
the reports submitted by a client while interacting with other providers will change the initial beliefs of a new provider . as we have seen in section [ buidingreputation ] ,
providers cheat less if they a priory expect with higher probability to encounter honest reporting clients .
a client that has once built a reputation for truthfully reporting the provider s behavior will benefit from cooperative trade during her entire lifetime , without having to convince each provider separately .
therefore the upper bound on the loss a client has to withstand in order to convince a provider that she is a commitment type , becomes an upper bound on the total loss a client has to withstand during her entire lifetime in the market .
how to effectively share the reputation of clients within the market remains an open issue .
correlated with this idea is the observation that clients that use our mechanism are motivated to keep their identity . in generalized markets where
agents are encouraged to play both roles ( e.g. a peer-2-peer file sharing market where the fact that an agent acts only as `` provider '' can be interpreted as a strong indication of `` double identity '' with the intention of cheating ) our mechanism also solves the problem signaled by related to cheap online pseudonyms .
the price to pay for the new identity is the loss due to building a reputation as truthful reporter when acting as a client . unlike incentive - compatible mechanism that pay reporters depending on the feedback provided by peers ,
the mechanism described here is less vulnerable to collusion .
the only reason individual clients would collude is to badmouth ( i.e. , artificially decrease the reputation of ) a provider
. however , as long as the punishment for negative feedback is not super - linear in the number of reports ( this is usually the case ) , coordinating within a coalition brings no benefits for the colluders : individual actions are just as effective as the actions when part of a coalition . the collusion between the provider and client
can only accelerate the synchronization of strategies on one of the ppe profiles ( collusion on a non - ppe strategy profile is not stable ) , which is rather desirable .
the only profitable collusion can happen when competitor providers incentivize normal clients to unfairly downrate their current provider .
colluding clients become _ malicious _ in this case , and the limits on the harm they can do are presented in section [ evilbuyers ] .
the mechanism we describe here is not a general solution for all online markets . in general retail e - commerce
, clients do nt usually interact with the same service provider more than once . as we have showed along this paper ,
the assumption of a repeated interaction is crucial for our results .
nevertheless , we believe there are several scenarios of practical importance that do meet our requirements ( e.g. , interactions that are part of a supply chain ) . for these , our mechanism can be used in conjunction with other reputation mechanisms to guarantee reliable feedback and improve the overall efficiency of the market .
our mechanism can be further criticized for being centralized .
the reputation mechanism acts as a central authority by supervising monetary transactions , collecting feedback and imposing penalties on the participants .
however , we see no problem in implementing the reputation mechanism as a distributed system .
different providers can use different reputation mechanisms , or , can even switch mechanisms given that some safeguarding measures are in place .
concrete implementations remain to be addressed by future work .
although we present a setting where the service always costs the same amount , our results can be easily extended to scenarios where the provider may deliver different kinds of services , having different prices .
as long as the provider believes that requests are randomly drawn from some distribution , the bounds presented above can be computed using the average values of @xmath186 , @xmath3 and @xmath187 .
the constraint on the provider s belief is necessary in order to exclude some unlikely situations where the provider cheats on a one time high value transaction , knowing that the following interactions carry little revenue , and therefore , can not impose effective punishments . in this paper , we systematically overestimate the bounds on the worst case percentage of false reports recorded by the mechanism .
the computation of tight bounds requires a precise quantitative description of the actual set of ppe payoffs the client and the provider can have in @xmath80 . and
pose the theoretical grounds for computing the set of ppe payoffs in an infinitely repeated game with discount factors strictly smaller than 1
. however , efficient algorithms that allow us to find this set are still an open question .
as research in this domain progresses , we expect to be able to significantly lower the upper bounds described in sections [ gtanalysis ] and [ buidingreputation ] .
one direction of future research is to study the behavior of the above mechanism when there is two - sided incomplete information : i.e. the client is also uncertain about the type of the provider .
a provider type of particular importance is the `` greedy '' type who always likes to keep the client to a continuation payoff arbitrarily close to the minimal one . in this situation
we expect to be able to find an upper bound @xmath188 on the number of rounds in which a rational client would be willing to test the true type of the provider .
the condition @xmath189 describes the constraints on the parameters of the system for which the reputation effect will work in the favor of the client : i.e. the provider will give up first the `` psychological '' war and revert to a cooperative equilibrium .
the problem of involuntary reporting mistakes briefly mentioned in section [ evilbuyers ] needs further addressing . besides false negative mistakes ( reporting @xmath29 instead of @xmath28 )
, normal clients can also make false positive mistakes ( report @xmath28 instead of the intended @xmath29 ) . in our present framework ,
one such mistake is enough ro ruin the reputation of a normal type client to report honestly .
this is one of the reasons why we chose a sequential model where the feedback of the client is not required if the provider acknowledges low quality .
once the reputation of the client becomes credible , the provider always rolls back the transactions that generate ( accidentally or not ) low quality , so the client is not required to continuously defend her reputation .
nevertheless , the consequences of reporting mistakes in the reputation building phase must be considered in more detail .
similarly , mistakes made by the provider , monitoring and communication errors will also influence the results presented here .
last , but not the least , practical implementations of the mechanism we propose must address the problem of persistent online identities .
one possible attack created by easy identity changes has been mentioned in section [ evilbuyers ] : malicious buyers can continuously change identity in order to discredit the provider . in another attack ,
the provider can use fake identities to increase his revenue .
when punishments for negative feedback are generated endogenously by decreased prices in a fixed number of future transactions , the provider can adopt the following strategy : he cheats on all real customers , but generates a sufficient number of fake transactions in between two real transactions , such that the effect created by the real negative report disappears . an easy fix to this latter attack
is to charge transaction or entrance fees .
however , these measures also affect the overall efficiency of the market , and therefore , different applications will most likely need individual solutions .
effective reputation mechanisms must provide appropriate incentives in order to obtain honest feedback from self - interested clients . for environments characterized by adverse selection , direct payments can explicitly reward honest information by conditioning the amount to be paid on the information reported by other peers . the same technique unfortunately does not work when service providers have moral hazard , and can individually decide which requests to satisfy . sanctioning reputation mechanisms
must therefore use other mechanisms to obtain reliable feedback .
in this paper we describe an incentive - compatible reputation mechanism when the clients also have a repeated presence in the market . before asking feedback from the clients
, we allow the provider to acknowledge failures and reimburse the price paid for service . when future transactions generate sufficient profit ,
we prove that there is an equilibrium where the provider behaves as socially desired : he always exerts effort , and reimburses clients that occasionally receive bad service due to uncontrollable factors .
moreover , we analyze the set of pareto - optimal equilibria of the mechanism , and establish a limit on the maximum amount of false information recorded by the mechanism .
the bound depends both on the external alternatives available to clients and on the ease with which they can commit to reporting the truth .
_ the probability that the client reports negative feedback on the equilibrium path of any pareto - optimal ppe strategy is zero . _
_ step 1_. following the principle of dynamic programming @xcite , the payoff profile @xmath99 is a ppe of @xmath80 , if and only if there is a strategy profile @xmath1 in @xmath22 , and the continuation ppe payoffs profiles @xmath190 of @xmath80 , such that : * @xmath102 is obtained by playing @xmath1 in the current round , and a ppe strategy that gives @xmath191 as a continuation payoff , where @xmath192 is the public outcome of the current round , and @xmath193 $ ] is the probability of observing @xmath192 after playing @xmath1 : + @xmath194 \cdot w_c(y ) \big ) ; \\ v_p & = ( 1-\delta)g_p(\sigma ) + \delta \big ( \sum_{y \in y } pr[y|\sigma ] \cdot w_p(y ) \big ) ; \end{split}\ ] ] * no player finds it profitable to deviate from @xmath1 : + @xmath195 \cdot w_c(y ) \big ) ; \quad \forall \sigma_c ' \neq \sigma_c \\ v_p & \geq ( 1-\delta)g_p \big ( ( \sigma_c,\sigma_p ' ) \big ) + \delta \big ( \sum_{y \in y } pr \big[y|(\sigma_c,\sigma_p')\big ] \cdot w_p(y ) \big ) ; \quad \forall \sigma_p ' \neq \sigma_p\\ \end{split}\ ] ] _ step 2 . _
take the ppe payoff profile @xmath196 , such that there is no other ppe payoff profile @xmath100 with @xmath101 .
let @xmath1 and @xmath190 enforce @xmath102 , and assume that @xmath1 assigns positive probability @xmath197 > 0 $ ] to the outcome @xmath198 .
if @xmath199 $ ] ( possibly equal to 0 ) , let us consider : * the strategy profile @xmath200 where @xmath201 is obtained from @xmath202 by asking the client to report @xmath28 instead of @xmath29 when she receives low quality ( i.e. , @xmath5 ) ; * the continuation payoffs @xmath203 such that @xmath204 and @xmath205 for @xmath57 .
since , the set of correlated ppe payoff profiles of @xmath80 is convex , if @xmath191 are ppe payoff profiles , so are @xmath206 . the payoff profile @xmath207 , @xmath208 is a ppe equilibrium profile because it can be enforced by @xmath209 and @xmath203 .
however , this contradicts our assumption that @xmath210 , so @xmath211 $ ] must be 0 . following exactly the same argument , we can prove that @xmath212 = 0 $ ] .
if no other ppe payoff profile @xmath100 can have @xmath214 , it must be that the continuation payoffs @xmath191 satisfy the same property .
( assume otherwise that there is a ppe @xmath215 with @xmath216 .
replacing @xmath217 in ( [ eq : step3 ] ) we obtain @xmath218 that contradicts the hypothesis ) . by continuing the recursion , we obtain that the client never reports @xmath29 on the equilibrium path that enforces a payoff profile as defined in step 2 .
pareto - optimal payoff profiles clearly enter this category , hence the result of the proposition .
since clients never report negative feedback along pareto - optimal equilibria , the only false reports recorded by the reputation mechanism appear when the provider delivers low quality , and the client reports positive feedback .
let @xmath42 be a pareto - optimal ppe strategy profile .
@xmath1 induces a probability distribution over public histories and , therefore , over expected outcomes in each of the following transactions .
let @xmath219 be the probability distribution induced by @xmath1 over the outcomes in round @xmath63 .
@xmath220 as proven by proposition [ prop : nozero ] .
the payoff received by the client when playing @xmath1 is therefore : since any ppe profile must give the client at least @xmath81 , ( otherwise the client is better off by resorting to the outside option ) , @xmath227 . by replacing the expression of @xmath228 , and taking into account the constraints on the probability of @xmath4 we obtain : | reputation mechanisms offer an effective alternative to verification authorities for building trust in electronic markets with moral hazard .
future clients guide their business decisions by considering the feedback from past transactions ; if truthfully exposed , cheating behavior is sanctioned and thus becomes irrational .
it therefore becomes important to ensure that rational clients have the right incentives to report honestly . as an alternative to side - payment schemes that explicitly reward truthful reports ,
we show that honesty can emerge as a rational behavior when clients have a repeated presence in the market . to this end
we describe a mechanism that supports an equilibrium where truthful feedback is obtained .
then we characterize the set of pareto - optimal equilibria of the mechanism , and derive an upper bound on the percentage of false reports that can be recorded by the mechanism . an important role in the existence of this bound
is played by the fact that rational clients can establish a reputation for reporting honestly . |
to generalize kl theory to magnetism , we need to first identify the theoretical constructs that underpin the theory of phonons and then seek analogs of these constructs in the magnon problem . the simplest description of phonons , that of the vibrations of balls connected to springs ( in the classical limit ) , is endowed with a number of theoretical constructs that shed much light on their behavior .
to begin with , this simplicity demands a specific form of their hamiltonian .
if we define the extension of spring @xmath3 to be @xmath4 , then the most general form for an ideal balls and springs classical hamiltonian is @xmath5 here @xmath6 is the @xmath7 component of the momentum of the ball labeled by @xmath8 , @xmath9 is the matrix inverse of the `` mass tensor '' @xmath10 , @xmath11 is the spring constant matrix and repeated indices are summed over . in the simplest setting , @xmath12 and @xmath13 are proportional to identity matrices . here
we leave them in the general form to aid our study of the structure of the theory and not its application .
the restriction of the hamiltonian to that of balls and springs therefore introduces two ( inverse ) metrics @xmath11 and @xmath9 in configuration space and momentum space respectively and identifies the ground state as satisfying constraints @xmath14 , @xmath15 . balls and springs are a toy - like simplification of a full phonon problem .
we will return to the justification of their use and the validity of the following results in the next section ( section [ sec : topologicalprotections ] ) . in the linearized phonon limit
, there is also a matrix @xmath16 and an associated topological invariant .
this matrix relates @xmath4 to the displacements of each atom @xmath17 from their equilibrium positions via @xmath18 .
this matrix also relates the forces @xmath19 to the tensions @xmath20 in each spring via @xmath21 .
the topological invariant associated with this matrix ( which we will call the maxwell - calladine index@xcite though it was discovered by kane and lubensky ) is @xmath22 where @xmath23 is the number of components @xmath24 , @xmath25 the number of sites , @xmath26 the number of springs ( bonds ) , @xmath27 denotes the dimension of the null space of @xmath28 and we have used @xmath29 by the fundamental theorem of linear algebra .
this quantity relates the topology data of the system , @xmath23 , @xmath25 and @xmath26 to the number of zero modes @xmath30 and number of states of self stress @xmath31 . here
a zero mode is a vector @xmath32 that satisfies @xmath33 and a state of self stress is a vector of tensions @xmath34 that satisfies @xmath35 .
this quantity is topological in the sense that it does nt depend on the metrics @xmath36 and @xmath9 and remains unchanged under any changes of the matrix @xmath16 .
so linearized balls and springs also have a natural topological invariant .
kane and lubensky argued that a dirac - like square - rooted equations of motion enables the study of phonons following the theory of topological insulators .
these equations of motion are defined by the matrix @xmath37 here i have taken the liberty to add a minus sign to kane and lubensky s matrix and re - inserted @xmath36 and @xmath9 which they set to identity matrices .
this is so that squaring this matrix produces @xmath38 where the upper left block defines the second order differential equation for the displacements @xmath39 the minus sign can be removed by multiplying @xmath40 by @xmath41 from the right , something we are free to do since @xmath42 commutes with the @xmath43 ( a feature noted by kane and lubensky ) .
this model helps us understand the importance of the topological index .
if , for periodic boundary conditions , @xmath44 then there must always be either a zero mode ( if @xmath45 ) or a self stress mode ( if @xmath46 ) . in the spectrum of @xmath40 there is no gap to the excitations .
however , if @xmath47 then we can always find an @xmath16 such that the gap disappears .
in such a case , if we open the boundary conditions we will invariably find that @xmath44 since we change the counting of the number of sites and bonds and so either special self stresses exist due to boundary conditions or special zero modes exist . in either case , there is an interplay between a gap , topology and boundary conditions as found in other topological systems such as the quantum hall effect or topological insulators .
finally , there is one more mathematical object we will find useful for the study of phonons . to study the zero modes of this problem directly
, we can follow ref . and
view the hamiltonian as energetically imposing constraints on the degrees of freedom . here
these constraints are simply @xmath48 and @xmath15 as noted above for if these conditions are met the hamiltonian vanishes . in his development of constrained hamiltonian mechanics ,
dirac pointed out that an important object in the study of constraints in phase space is the constraint matrix which here takes the form : @xmath49 where @xmath50 is the usual poisson bracket of classical mechanics .
a vector in the null space of this matrix is then either associated with a redundant constraint ( when there are more constraint functions than necessary to constrain the variables ) or with a zero mode coordinate that has no conjugate variable in the space of zero modes ( gauge coordinate ) .
if we expand the spring extensions @xmath4 in terms of the displacements of the atoms from their equilibrium positions @xmath17 as @xmath51 then the constraint matrix takes the form @xmath52 remarkably , in this form , @xmath53 is similar to the `` square - rooted '' hamiltonian @xmath40 discussed above .
if we work in kane and lubensky s units where @xmath54 and @xmath55 they are actually the same .
we also see that the redundant constraints are associated with self stress modes living in the null space of @xmath56 and the gauge coordinates with the zero modes living in the null space of @xmath16 .
so the eigenvalue problem associated with dirac s constraint matrix here is closely related to the dirac - like square - rooted phonon hamiltonian !
now the relationship between all of these mathematical objects , the metrics , the hamiltonians , the @xmath16 matrix , and the constraint matrix are not so easily understood .
they all clearly derive from the balls and springs construction . but
how is the topological index dependent on the metrics ?
how is the constraint matrix @xmath57 related to the square - rooted hamiltonian @xmath40 ? in what way is the topology protected ? can these constructions be applied to other systems that are not balls and springs ? to address these questions , we need an abstract theoretical framework that identifies all of these mathematical object from a simpler set of constructs . remarkably , such a framework is provided by extending the model s phase space to include a set of fermion degrees of freedom @xmath58 and @xmath59 that we will call phonino modes and imposing supersymmetry . from this vantage point
we will then look down on all the mathematical objects discussed previously and see their relationships to each other .
the formal nature of the supersymmetry will then also allow us to connect two seemingly different problems : balls and springs and quadratic spin models .
we then just need to determine if the connection is merely formal or whether it predicts new physical phenomena for either balls and springs or magnetic systems .
a hamiltonian that is a quadratic form in constraint functions that define its ground state has a natural supersymmetry . to construct this supersymmetric model , we begin by defining the supersymmetric charge by the product of a constraint function and a new degree of freedom : @xmath60 this looks like a lagrange multiplier term in an action meant to impose the constraints @xmath48 and
@xmath15 where @xmath59 and @xmath58 are the lagrange multipliers .
however , here we view these as degrees of freedom that satisfy their own poisson bracket relations .
indeed we can construct the supersymmetric hamiltonian using the poisson bracket @xmath61 where @xmath62 denotes a poisson bracket .
for this equation to be true , it is necessary for @xmath58 and @xmath59 to be grassmann numbers so that @xmath63 is grassmann odd and @xmath64 does not vanish by the normal antisymmetry of poisson brackets .
we also used @xmath65 .
if we choose the remaining poisson brackets to be @xmath66 then we obtain the simple relation @xmath67 where @xmath68 is the first term and the second and third terms make up @xmath69 of eq .
[ eq : hphonon ] .
this defines the poisson bracket to be@xcite @xmath70 which is symmetric if @xmath71 and @xmath72 are both grassmann odd and antisymmetric otherwise .
finally , we see that @xmath73 so the two observables form a closed superalgebra .
so we can promote the phonon problem to a supersymmetric problem of bosonic phonons and fermionic phoninos where the phoninos satisfy a clifford algebra with metrics @xmath11 and @xmath9 and the phonino hamiltonian is determined by the constraint matrix @xmath74 where @xmath53 is given by eq .
[ eq : phononcmatrix ] . note , here we have chosen to study phonons classically using grassmann numbers . in this sense
we follow one of the original motivations to introduce grassmann numbers into classical mechanics : to construct simpler - than - field - theory supersymmetry models@xcite .
the use of classical mechanics here , however , is just to connect with the history of theories of balls and springs and is not necessary as we will discuss section [ sec : topologicalprotection ] .
furthermore , our analysis of the model presented below will be informed by the use of supersymmetry in quantum mechanics ( see ref . for a review ) , because it is another simpler - than - field - theory supersymmetry model . however , unlike quantum mechanics , the above model is a many - body supersymmetric lattice model , one not so different from the lattice model of a supersymmetric topological superconductor studied in ref . .
in the linearized limit the two hamiltonians become @xmath75 and @xmath76 where we used @xmath77 . in this limit then , the phonons and the phoninos are decoupled and the hamiltonians are quadratic .
we therefore have a model we can apply to real systems since it contains the correct phonon eigenvalue problem .
it is not so obvious , however , what we can learn by introducing the phoninos . to settle this , we need to directly study the different eigenvalue problems .
to study the eigenvalue problems of @xmath78 , we need to solve the corresponding equations of motion .
the first order in time differential equations for the phonons are @xmath79 and for the phoninos are @xmath80 remarkably , the phonino equations of motion are the square - rooted hamiltonian matrix @xmath40 introduced by kane and lubensky . the constraint matrix @xmath53 and @xmath40
are therefore related in the supersymmetric model by one entering the hamiltonian quadratic form and the other by the corresponding equations of motion .
we will also find it useful to work with the second derivative in time equations of motion : @xmath81 from these , we see that , @xmath17 , @xmath82 and @xmath59 all obey the same equations of motion . @xmath58 obeys a distinctly different equation .
we also can determine that each equation has two types of eigenmodes .
for the @xmath17-type equations , all modes either have zero eigenvalue and are in the null space of @xmath16 or they have finite eigenvalues .
there are no modes with zero eigenvalue that satisfy @xmath83 .
this is because such a mode would satisfy @xmath84 with @xmath85 and violate the assumption that @xmath36 is a positive definite matrix . for a similar reason ,
the @xmath58 equation has two modes types , those with zero eigenvalue in the nullspace of @xmath56 and those with finite eigenvalues .
we can of course proceed by solving these equations separately , but this would not give us any insight into how the phoninos can help us understand the phonons .
instead , lets proceed by studying supersymmetry transformations and reduce the problem to solving only the parts not related by supersymmetry .
the supersymmetric charge provides us with a map between the phase space observables with an even number of phonino coordinates ( grassmann even ) and those with an odd number ( grassmann odd ) .
the single particle phonon modes which are linear combinations of @xmath17 and @xmath6 are in the even group while the single particle phonino modes which are linear combinations of @xmath59 and @xmath58 are in the odd group .
the map @xmath86 we can construct from @xmath63 sends an observable from the even sector to the odd sector or vice versa by inserting it into the location of the @xmath87 .
this map has a special relationship with the equations of motion : time evolution of the mapped observable is the same as the unmapped observable .
for example , @xmath88 is a phonino observable that obeys the equations of motion @xmath89 which is the equations of motion for @xmath17 .
similarly , the phonon observables @xmath90 , @xmath91 , @xmath92 obey the @xmath6 , @xmath59 and @xmath58 equations respectively .
computing these poisson brackets explicitly , we see that they are @xmath93 so it would seem that supersymmetry is too powerful a symmetry , that the phonino problem is equivalent to the phonon problem ( and therefore uninteresting ) . to study these relationships in more detail , lets now take linear combinations of phase space observables .
if @xmath94 is an eigenmode of the @xmath6 equation then it obeys @xmath95 now if @xmath96 so the @xmath97 is in the ( left ) nullspace of @xmath56 , then it is an eigenvector with frequency @xmath98 .
if this is not the case , it is an eigenvector with a finite frequency as discussed above .
now lets map this to a linear combination of @xmath58 s using @xmath63 : @xmath99 if @xmath97 therefore corresponds to a zero mode , it gets annihilated by this map . only the finite frequency modes
therefore actually pass from the @xmath6 eigenmode problem to the @xmath58 eigenmode problem .
however , for the finite mode case , notice that the mapped observable is a linear combination whose coefficients obey @xmath100 so it is a ( left ) eigenvector of the @xmath58 differential equation with the same eigenvalue it had in the @xmath6 differential equation before it was mapped . mapped
observables therefore carry their eigenvalue with them . extending this study to each of the four differential equations and breaking their set of eigenvectors into two groups , the finite and the zero eigenvalue modes , we have constructed fig .
[ fig : phononqmap ] that graphically depicts all the different relationships . in it
an arrow represents a function mapping the set of eigenmodes at its tail to the set of eigenmodes at its head . if two different paths exist between any two sets , one going from one set to the other and the other going the reverse direction , then the two sets are isomorphic and in particular have the same number of eigenmodes . in this way
, we see that the number of finite energy eigenmodes are the same for all differential equations , just their zero modes are different . ''
is the supersymmetry map .
notice @xmath63 does not map zero mode eigenstates from @xmath6 to @xmath58 and from @xmath58 to @xmath17 .
since all arrows represent maps of unique elements to unique elements , if a path of arrows exists from one set of modes to another and another path exists from that set of modes back to the first then the two sets of modes are isomorphic . for finite modes ,
any two sets of eigenstates have such paths relating them so they are all isomorphic to each other . for zero modes , only the @xmath17 , @xmath59 and @xmath6 sets of eigenstates
are isomorphic .
the @xmath58 set is different and hence the phonino problem has a different set of zero modes from the phonon problem .
note this figure follows the rules of an olog@xcite.,title="fig : " ] '' is the supersymmetry map .
notice @xmath63 does not map zero mode eigenstates from @xmath6 to @xmath58 and from @xmath58 to @xmath17 .
since all arrows represent maps of unique elements to unique elements , if a path of arrows exists from one set of modes to another and another path exists from that set of modes back to the first then the two sets of modes are isomorphic . for finite modes ,
any two sets of eigenstates have such paths relating them so they are all isomorphic to each other . for zero modes , only the @xmath17 , @xmath59 and @xmath6 sets of eigenstates are isomorphic .
the @xmath58 set is different and hence the phonino problem has a different set of zero modes from the phonon problem .
note this figure follows the rules of an olog@xcite.,title="fig : " ] now , with the above analysis of the eigenmodes , we are in a position to understand the topological index associated with a supersymmetric theory . as pointed out by witten ( and often called the `` witten index '' ) the total number of boson modes minus the total number of fermion modes ( @xmath101 where @xmath102 for a boson mode , and @xmath103 for a fermion mode ) is a topological invariant . applying this to the single particle sector we have @xmath104 but if we break down the count of zero modes into the four groups of two types of modes , we have @xmath105 where we used fig . [
fig : phononqmap ] to establish that the number of finite modes @xmath106 is the same between all modes types while @xmath107 is the same only between @xmath17 , @xmath59 and @xmath6 and is in general different from the number of @xmath58 zero modes , @xmath108 . to derive these results
, we heavily relied on the positive definiteness of @xmath9 and @xmath36 for if this were not the case , more modes could fail to be mapped by @xmath63 .
so the witten index of supersymmetry , when applied to the single particle sector valid for the linearized theory , is identical to the maxwell - calladine index but its derivation is directly built on eigenvalue problems rather than directly on the properties of the matrix @xmath16 .
the witten index is useful in a supersymmetric theory because it dictates whether there must be a zero mode . if @xmath109 there must be @xmath110 bosonic ( phonon ) zero modes . if @xmath111 there must be @xmath112 fermionic ( phonino ) zero modes which here must be self stress modes .
so long as @xmath113 then , supersymmetry can exist in the `` ground state '' for there is then a zero mode that satisfies @xmath114 ( in the quantum language ) .
however , if @xmath47 then without explicitly breaking supersymmetry one can remove all zero modes .
then there are no states satisfying @xmath114 and supersymmetry is spontaneously broken in the ground state . remarkably , an isostatic lattice has both @xmath47 for periodic boundary conditions and all zero modes gapped ( @xmath115 ) .
so isostatic lattices are an example spontaneous susy breaking . finally , we need to comment on the question of how the above supersymmetry is related to the quantum mechanics - like supersymmetry of the square rooted equations of motion ( eq . [ eq : sqrtedeom ] ) pointed out by kane and lubensky .
this symmetry is a supersymmetry between @xmath59 modes and @xmath58 modes ( both of which are fermionic here ) and has an associated topological index which is numerically the same index as the single particle witten index discussed above and the maxwell - calladine index .
so the phonon problem appears to have two different supersymmetries , one at the full phase space level that we discuss above and the other at the level of the equations of motion of the linearized fermions that related different fermion modes to each other ( quantum mechanics - like susy need not always relate bosonic modes to fermionic modes ) . in summary
, the supersymmetry model has brought all mathematical objects mentioned so far under one umbrella .
the supersymmetric theory has the phonon hamiltoinan , the two metrics @xmath9 and @xmath36 play a central role as they determine the commutation relations of the fermions , the phonino hamiltonian is determined by the constraint matrix @xmath53 and its corresponding equations of motion is the square - rooted hamiltonian @xmath40 .
finally , the topological index is determined by the supersymmetry relations between the various eigenmodes .
lastly , it suggests that the specialness of the balls and springs description is the requirement that it is supersymmetric .
if we break this specialness , if we explicitly break the supersymmetry , then most of the mathematical objects contained in the theory loose their meaning . in this way
, i argue that the topological structure of isostatic lattices identified by kane and lubensky is susy protected .
the previous results show that models of phonons built on constraint functions like the extension @xmath4 of a spring from its equilibrium length have a topological index tied to their equations of motion by a supersymmetric relationship with phoninos .
this index guarantees the existence of zero modes from topology ( even in the absence of a continuous symmetry ) much like goldstone s theorem guarantees the existence of zero modes from spontaneous symmetry breaking .
here we discuss the question of to what extent this topological property is robust . for the most part , we worked in the classical limit where phonons are actually lattice vibrations .
if we quantize it , then we convert poisson brackets to commutators or anticommutators such as @xmath116 = i\delta_{i\alpha}^{j\beta}$ ] and @xmath117 . quantizing the supersymmetric charge
@xmath118 then gives the supersymmetric hamiltonian @xmath119 this is identical in form to the classical case .
a similar correspondence between the non - linear forms of the classical and quantum model also holds .
so the main results of this paper hold in both the classical and quantum forms and are not restricted to semi - classical or classical regimes .
since the metrics @xmath36 and @xmath9 entering the phonon problem are not present in the respective topological indices , the topological protection holds if these undergo a distortion so long as they remain positive definite matrices . in particular , the topological index holds if we stretch and distort the lattice as was used extensively in the discovery of weyl phonon modes in ref . .
let us first address this question for the non - linear supersymmetry models before returning to the problem of purely non - linear phonons in the absence of phoninos . in the presence of non - linearities , the eigenmodes of the system are no longer linear combinations of single particle observables .
instead , they become a linear combination of all multiparticle observables . as a result , the witten index restricted to the single particle sector
looses its meaning but the full index @xmath120 still holds because @xmath63 still maps multiparticle bosonic observables to a multiparticle fermionic observables . to get a sence of this full set of observables
, we have considered the two - particle sector .
this consists of bosonic observables of types @xmath121 , @xmath122 , @xmath123 , @xmath124 , @xmath125 and @xmath126 and fermionic observables of types @xmath127 , @xmath128 , @xmath129 and @xmath130 . counting all of these two different two particle observables , noting that squares of grassmann observables vanish ( i.e. @xmath131 ) i obtain after simplifying @xmath132
if @xmath133 as in the isostatic lattice case , we see this also vanishes .
since the index for the one - particle and two - particle sectors vanishes , likely the total sum over all sectors relevant for the many body non - linear problem also vanishes .
so the topological protection remains in effect for the full non - linear supersymmetric problem . with this technical extension ,
we see that the non - linear models are also topological .
since they consist now of majorana fermions coupled to phonons , they could in principle be realized in a carefully engineered superconductor . in practice , however , they may prove to be more useful as a means of gaining theoretical insight into `` symmetry protected topological order '' in the presence of interacting fermions which does not make use of k - theory@xcite , chern - simons quantum field theory@xcite or supercohomology@xcite .
unfortunately , the salient feature of the non - linear problem is that the phonons are now coupled to the phoninos
. we can no longer think about the added fermionic degrees of freedom as a device to construct a topological index .
so , if the fermions are absent the supersymmetry is lost and with it the topological protection .
this statement , however , is about the microscopic physics . it may very well be that the full non - linear phonon problem has the same low energy long distance physics as the corresponding supersymmetric non - linear phonon . after all , it is the physics close to the ground state which we used to build the model of balls and springs from constraint functions .
it is therefore possible that the susy breaking terms are irrelevant in the renormalization group sense .
if so , the topological protection may nevertheless emerge in the low energy limit .
the models discussed in this paper are all `` toy '' models in that they correspond to useful simplifications of the actual phonon microscopic physics but in any real system , there will be perturbations that are not of the simple form .
for example , the general quadratic phonon potential @xmath134 is not that of balls and springs ( @xmath135 ) .
the supersymmetry can therefore be thought of as a symmetry emerging from the ` toy - ness ' of the model .
we can speculate on what this might mean .
the extent to which the toy model captures the full physics of a real material is likely the extent to which the supersymmetry will be obeyed .
in particular , if supersymmetry demands gappless phonon edge modes in a system with gapped bulk phonon modes , these edge modes would likely be gapped by perturbations violating the toy - like property of balls and springs and the size of the gap is likely related to the strength of these non - toy - like perturbations .
we can extend the above phonon discussion to magnons ( or other similar systems ) by recognizing its essential ingredient : constraints .
the phonon problem is built on the local - in - configuration space condition that the extensions @xmath4 vanish in the ground state .
since a ground state also requires a vanishing momenta @xmath6 , we see that the phonon problem is a quadratic form in the set of local - in - phase space constraints on the ground state .
to extend to magnons , we therefore need models built around local - in - phase space constraints that define a ground state magnetic ordering pattern .
perhaps the simplest example of such a constraint in a magnetic system is the vanishing total ( classical ) spin of a nearest neighbor bond in the nel state of the square lattice antiferromagnet .
each nearest neighbor bond @xmath3 has a pair of sites @xmath136 that satisfies @xmath137 .
here each @xmath138 is a dynamical spin vector satisfying angular momentum poisson brackets @xmath139 and lives in a spherical @xmath140-dimensional phase space .
such a condition is directly analogous to a motionless unstreatched spring with vanishing extension @xmath15 and momenta @xmath14 , @xmath141 .
so we can follow our phonon discussion and use the condition @xmath142 to write down the susy charge @xmath143 and susy hamiltonian @xmath144 choosing @xmath145 a positive definite symmetric matrix makes the first term a typical quadratic spin model with @xmath142 for a ground state ( e.g. the case @xmath146 is just the heisenberg model ) .
this makes the @xmath147 grassmann variables obeying a clifford algebra just like we found for the phonon problem .
additionally , @xmath148 follows from @xmath149 above and defines the fermion hamiltonian again as dirac s constraint matrix .
we can then expand about a ground state to linear order using @xmath150 with @xmath151 the phase space coordinates representing deviations of the spin on site @xmath8 from its ground state value .
this renders @xmath152 constant and decouples the two terms in @xmath78 .
linearizing allows us to demote the magninos described by @xmath147 to fictitious particles whose use is to reveal the implications of the supersymmetry just like the case of the phoninos .
however , the equations of motion of these magninos does not give us the `` square - rooted '' hamiltonain for it is @xmath153 which , when squared , is not the magnon equations of motion .
so the structure of the square - rooted hamiltonian and its associated quantum mechanics - like supersymmetry does nt apply here .
thus , by starting from the ground state condition @xmath142 we obtain the same structure as the phonon problem which presumably has a witten index @xmath110 that can tell us about its zero modes without solving the equations of motion .
the above nel state bond constraint @xmath154 is an application of the susy phonon problem to unfrustrated antiferromagnets .
it readily generalizes to the ferromagnetic case which has a vanishing difference of spin vectors on each bond ( @xmath155 ) provided we draw arrows on each bond to specify which of the two sites of nearest neighbor bond @xmath3 enters @xmath154 with a negative sign .
it also readily generalizes to other bi - partite lattices such as the cubic lattice and the honeycombe lattice .
writing down the topological index expected for these generalizations will therefore allow us to determine if an analog of isostatic phonons exists and also how we may use it to study the loss of magnetic order . for each case
, we can guess the form of the topological index @xmath156 the last line here is not a guess , it follows from the previous line by the fundamental theorem of linear algebra as before . presumably , if @xmath157 , there must be a finite number of magnon zero modes and if it is negative a finite number of magnino modes .
we will discuss the meaning and correctness of this statement below .
but assuming it is correct for the present discussion , we see that essentially all unfrustrated magnets have @xmath158 . only dimerized systems which have two sites for each bond or trimerized systems with _ two bonds _ for each trimer have non - negative @xmath159 ( see fig .
[ fig : unfrustratednu ] ) .
the best case with connected sites is the one dimensional chain with @xmath160 and @xmath161 for periodic boundary conditions .
a similar best case holds for bond percolation on the square lattice where at threshold @xmath162@xcite .
so for unfrustrated magnets , the topological structure imposed by supersymmetry does nt demand any magnon zero modes . becomes increasingly negative with increasing chain length going negative for four or more spins , essentially all unfrustrated magnetic systems will have a negative topological index .
this explains the general robustness of magnetic order in many magnetic materials . ]
the above derivation of the magnetic maxwell - calladine index was mostly a guess . to see that it is correct
, we need to work out the supersymmetry relations between eigenmodes and the witten index as we did for the phonon problem .
this is a more complicated problem than the phonon case due to the existence of two types of zero modes@xcite ( canonical and gauge ) .
but this just requires a little more organization and the following proceeds just like the phonon case .
the equations of motion for both bosonic and fermionic problems are @xmath163 where @xmath164 is the poisson bracket tensor with @xmath165 the two dimensional levi - civita tensor and we expanded the constraint matrix to leading order ( where @xmath166 ) .
lets identify first the types of bosonic eigenmodes , then the types of fermionic eigenmodes and then see how they map into each other by sypersymmetry . for the @xmath167 equations of motion
, we can break up the set of eigenmode observables @xmath168 into four types of modes . 1 . finite frequency modes obeying @xmath169 with @xmath170 ; 2 .
gauge zero modes that are eigenmodes with @xmath98 but whose observable @xmath171 commutes with all other observables corresponding to other @xmath98 eigenmodes .
these obey @xmath172 and @xmath173 where @xmath174 is any vector satisfying @xmath175 .
3 . canonical zero modes that are eigenmodes with @xmath98 that are not gauge zero modes .
these obey @xmath172 and @xmath176 for some @xmath174 satisfying @xmath175 .
non - eigenmodes of the equations of motion that can arise because the eigenvalue problem here is a non - symmetric matrix that does not guarantee a complete set of eigenvectors .
these turn out to be the modes conjugate to the gauge zero modes .
we will therefore call them `` conjugate gauge modes '' .
these modes are best understood by choosing a canonical basis that pairs them up into position @xmath177 and momentum @xmath178 observables
. then each pair has hamiltonian @xmath179 in this light , finite frequency modes have @xmath180 , @xmath181 , canonical zero modes have @xmath182 , @xmath183 and @xmath184 and gauge modes have @xmath185 , @xmath183 . for the @xmath147 equations of motion
, we can break up the set of eigenmodes @xmath186 into : * finite frequency modes obeying @xmath187 ; * self field modes obeying @xmath188 ; * magnino gauge modes that do not satisfy @xmath189 but are eigenmodes with @xmath98 .
self field modes are modes that correspond to a choice of local magnetic fields on the bonds @xmath190 that contribute no zeeman energy @xmath191 they identify a redundancy in the set of functions @xmath154 when viewing them as constraints on the ground state configuration .
the `` magnino gauge modes '' are not gauge modes in the sense of zero modes whose conjugate is not a zero mode .
their name here comes from their use in counting gauge modes : dirac found@xcite there is a gauge mode for every eigenvector in the nullspace of the constraint matrix that does not correspond to a redundant constraint ( i.e. is not self field mode ) .
now with all these modes identified , we can pass them through the supersymmetry map @xmath86 and see how they are related by supersymmetry .
the result is presented in fig .
[ fig : fmagnonsqmap ] . since no two eigenmodes mapped by @xmath86 maps to the same eigenmode
, we must have : * the number of finite frequency magnon modes @xmath192 is equal to the number of finite frequency magnino modes ; * the number of magnon gauge modes @xmath193 is equal to the number of magnino `` gauge '' modes . in addition
, we see that the supersymmetry map gives us a direct explanation of dirac mode counting : magnino gauge modes map to magnon gauge modes under supersymmetry and the conjugate to this magnon gauge mode maps back to the magnino gauge modes .
this defines an isomorphism between magnon and magnino gauge modes and so the number of each must be the same .
the supersymmetry map also verifies our guess for the single particle witten index / magnetic maxwell - calladine index for the above conditions on the number of modes in each set implies the difference between the number of magnon and magnino modes is @xmath194 where @xmath195 is half the number of zero modes with a conjugate pair that is also a zero mode and @xmath196 is the number of self field modes . and so our guess of eq .
[ eq : aindex ] is indeed correct with a positive @xmath110 demanding @xmath197 and a negative @xmath110 demanding @xmath198 .
we can also determine the topological another way .
we see from the above discussion that @xmath199 ( dimension of the constraint surface @xmath200 in phase space ) and @xmath201 since these null spaces define the modes that do not get mapped under supersymmetry .
hence , the witten index in the magnon problem also obeys @xmath202 . , this figure is also an olog@xcite . ]
so our guess for the topological index is correct but this in turn implies that unfrustrated magnons are always far away from the analog of mechanical collapse expected for @xmath45 ( i.e. they all have a negative @xmath110 ) .
so , to explore the physics of the kane - lubensky theory in magnetic systems , we need to generalize beyond unfrustrated magnetic systems . but
before doing so , we turn to an illustrative example of the above magnon and magnino eigenvalue problems : the simple ferromagnet on the square lattice
. we can write the hamiltonian for the square lattice ferromagnet heisenberg model as @xmath203 where @xmath146 , @xmath3 labels nearest neighbor bonds @xmath204 and @xmath24 the spin vector components .
the ground state therefore satisfies @xmath205 on each horizontal bond and @xmath206 on each vertical bond .
up to global spin rotations , this uniquely selects the ground state to be the uniform state . to linearize around a ground state , let s choose the case @xmath207 , @xmath208 .
expressing the spin vectors using canonical coordinates @xmath209 with @xmath210 so that @xmath211 we can linearize by expanding in @xmath212 and @xmath213 .
this gives @xmath214 .
similarly , we can expand the constraint @xmath215 and identify @xmath216 and @xmath217 with @xmath218 indexing to which canonical variable @xmath167 refers .
so we find @xmath219 where the sign depends on which spin @xmath220 refers to in the expression @xmath221 . now since the ground state is periodic with the lattice , we may fourier transform . to do this ,
let @xmath222 with @xmath223 or @xmath224 denoting the center of each bond and @xmath225 denoting the site of the lattice . then @xmath226 and we see its translational symmetry by just shifting both @xmath227 and @xmath228 by @xmath229 and for some lattice vector @xmath229 .
we can then expand @xmath230 in fourier components by @xmath231 where @xmath232 is the number of unit cells .
this introduces the 6x2 matrix @xmath233 and allows us to express the magnon hamiltonian as @xmath234 multiplying the matrices out for the nearest neighbor heisenberg model ( @xmath235 a 6x6 identity matrix ) we find @xmath236 where @xmath237 is the 2x2 identity matrix giving a quadratic dispersion for small @xmath238 as expected .
now we can identify mode types as finite frequency , magnon gauge , conjugate to magnon gauge and magnon canonical zero mode . at @xmath239
we find a zero mode with @xmath240 .
it is therefore of the form of eq .
[ eq : modeform ] with @xmath241 , @xmath242 and @xmath243 .
hence it is a canonical zero mode . for finite @xmath238
, we see that hamiltonian is of the form of eq .
[ eq : modeform ] with @xmath244 and @xmath245 finite .
hence these are finite frequency magnon modes .
there are unfortunately no magnon gauge modes are conjugate gauge modes to illustrate this point .
the magnino hamiltonian takes a similar form .
it is @xmath246 where @xmath247 is the antisymmetric poisson bracket tensor .
now @xmath248 is a @xmath249 antihermetian matrix : @xmath250 choosing @xmath235 a 6x6 identity matrix so that @xmath251 obey canonical commutation relations then allows us to diagonalize this matrix .
the result is four zero eigenvalues , @xmath252 and @xmath253 .
hence the finite frequency modes have exactly the same eigenvalues .
again we can distinguish the different mode types though here we havent expressed a simple canonical form for the diagonalized fermion hamiltonian .
we do nt have eq .
[ eq : modeform ] to help guide us .
but we can work directly with the definitions of the modes .
clearly finite frequency modes magnino modes correspond to the eigenvalues @xmath252 and @xmath253 for @xmath254 and there are exactly the same number of these as finite frequency magnon modes .
the four zero eigenvalues and the case @xmath255 fall into the null space of @xmath256 and are `` self field modes '' .
there are no zero frequency modes not in the null space of @xmath257 so there are no `` magnino gauge modes '' consistent with the absence of magnon gauge modes discuss above .
perhaps one way to overcome the ubiquitous negative @xmath110 of unfrustrated magnets is to generalize to other degrees of freedom . for this to improve upon the unfrustrated case
, we need to increase the number of degrees of freedom on each site compared to the number of constraints on each bond .
this seems difficult for magnetic systems for each site naturally has its own constraints ( such as unit length spin vectors ) in addition to the bond constraints .
for example , consider @xmath258 quantum rotors each characterized by @xmath258 spin unit vector @xmath259 , ( @xmath260 , @xmath261 ) and angular momenta @xmath262 conjugate to @xmath263 ( see sachdev s
book@xcite ) . in this case
@xmath264 and @xmath265 in the ground state so we can define @xmath266 .
such a @xmath63 leads to @xmath267 with fermions again decoupling in the linearized limit .
but now our guess for the topological index is @xmath268 which is typically even more negative than @xmath159 ( and the situation does nt improve for @xmath269 rotors ) .
but notice that here we kept the same number of phase space degrees of freedom on each site and just transferred one of the bond constraints to a site constraint .
thus it is even more difficult to introduce magnon zero modes for quantum rotors than for quantum spins and they do not seem to be a viable solution . let us then turn to changing the constraint functions entirely .
lets replace the vanishing bond constraint @xmath142 with a vanishing total spin on a plaquette constraint @xmath270 .
such a constraint arises in the nearest neighbor triangular , kagome , pyrochlore antiferromagnets and the @xmath271-@xmath272 square lattice antiferromagnet with @xmath0 .
all of these cases fall in to the category of frustrated magnets .
for example , the triangular lattice heisenberg model can be written @xmath273 where @xmath274 labels each triangle of the triangular lattice and the @xmath271-@xmath272 square lattice antiferromagnet at @xmath275 can be written @xmath276 we can again guess the topological index . here
it is @xmath277 where @xmath278 is the number of plaquette constraints ( and fermion grassmann numbers ) .
this gives for @xmath232 unit cells and periodic boundary conditions @xmath279 for the triangular lattice ( 2 triangles , 1 site in unit cell ) , @xmath280 for the @xmath281 square lattice case ( 1 square , 1 site in unit cell ) , @xmath282 for the kagome case ( 2 triangles , 3 sites in unit cell ) and @xmath283 for the pyrochlore lattice ( 2 tetrahedrons , 4 sites in unit cell ) .
so frustration enables a wide variety of models to explore the impact of the full range of @xmath110 values on the ordering tendencies of the spins .
we next work out the special @xmath47 kagome case as a concrete example and use it to clarify a number of the above general statements . on the kagome lattice ( and indeed many highly geometrically frustrated magnets@xcite ) the hamiltonian
can be written @xmath284 where @xmath285 is the total spin on triangle @xmath286 .
for the case ,
@xmath287 , this model reduces to the nearest neighbor heisenberg model on a lattice of corner sharing triangles such as the kagome lattice and hyperkagome lattice .
we will therefore assume @xmath288 has positive definite eigenvalues and defines a metric .
this model has the feature that the ground states all satisfy the local condition @xmath200 . using this constraint
, we can again write down a supersymmetric charge @xmath289 and compute the corresponding supersymmetric hamiltonian @xmath290 so we need to choose @xmath291 to make the first term @xmath292 .
the second term is the magnino hamiltonian and like the phonino hamiltonian is determined by the constraint matrix ( @xmath293 ) .
hence , we obtain the same structure as we did for the unfrustrated case .
we just replace the index @xmath3 with @xmath274 .
we can now compute the topological index for kagome antiferromagnets .
after linearizing as before the analog of the magnetic maxwell - calladine index is now @xmath294 this is different from the unfrustrated case only in replacing @xmath26 the number of bonds with @xmath295 the number of triangles . for periodic boundary conditions ,
we see that @xmath296 where @xmath232 is the number of unit cells considered , @xmath297 is the number of sites within a unit cell and @xmath298 is the number of triangles within a unit cell .
we additionally see this holds for other kagome lattices like the hyperkagome lattice of na@xmath299ir@xmath2o@xmath300@xcite in three dimensions which has @xmath301 , @xmath302 .
so the kagome family of lattices all have @xmath47 .
given that @xmath47 for periodic boundary conditions , lets see what the index demands for various open boundary conditions .
for the two finite kagome clusters shown in fig .
[ fig : kagomeedgestates ] , we have @xmath303 and @xmath304 .
so either type of boundary condition will have @xmath305 and demand zero modes . presumably , this holds for any boundary condition that still allows us to write the hamiltonian in the general form of eq .
[ eq : hfrustratedmagnons ] .
however , other boundary conditions may include a bond constraint function @xmath154 in place of a triangle constraint function @xmath306 and we would need to rethink the problem and be sure these constraints really do define ground state configurations ( i.e. it could be that there are no spin configurations that satisfy all the constraints ) . in these cases , it may be that @xmath111 for such boundaries given that we found bond constraints negatively contribute to @xmath110 so that @xmath110 does nt demand any zero modes .
hence , it may depend on the nature of the boundary conditions as to whether zero modes are expected .
however , there exist boundary conditions where the topological index will demand the existence of zero modes even in the absence of any spin rotational symmetry breaking ( as it does in systems with @xmath45 for periodic boundary conditions such as the nearest neighbor pyrochlore antiferromagnet ) . .
a ) has dangling triangles and @xmath307 , b ) has no dangling triangles and @xmath308 .
it appears that @xmath45 for any open boundary conditions and only reaches zero for periodic boundary conditions .
thus we can always gap out all bulk modes with a suitable choice of the matrix @xmath309 but not so at the edge . a magnon gauge or canonical
zero mode must exist on the edge so long as the magnon hamiltonian is of the form eq .
[ eq : hfrustratedmagnons].,title="fig : " ] .
a ) has dangling triangles and @xmath307 , b ) has no dangling triangles and @xmath308 .
it appears that @xmath45 for any open boundary conditions and only reaches zero for periodic boundary conditions .
thus we can always gap out all bulk modes with a suitable choice of the matrix @xmath309 but not so at the edge . a magnon gauge or canonical
zero mode must exist on the edge so long as the magnon hamiltonian is of the form eq .
[ eq : hfrustratedmagnons].,title="fig : " ] the above results show that indeed we can extend the supersymmetry of the linearized phonon problem to both unfrustrated and frustrated magnons .
in addition , we find that the kagome family is an analog of `` maxwell lattices''@xcite .
it has @xmath47 for periodic boundary conditions and @xmath45 at least for certain open boundary conditions .
if we gap out all modes , with supersymmetry allowed perturbations , we would then have a magnetic analog of an isostatic lattice ( a maxwell lattice with no zero modes ) .
let us then turn to illuminating some of the new physics that is transported from the theory of phonons into the theory of magnetism by our formal developments .
perhaps most striking is the existence of a new phase of magnetism which is on the verge of magnetic collapse that we will here refer to as isostatic magnetism .
we can understand the existence of this phase in simple terms as shown in fig .
[ fig : isomagnetism ] . first consider the three different mechanical structures , the square , the square with one diagonal beam and the square with two diagonal beams we see three different forms of rigidity : floppy , fragile , and firm .
the square is floppy because it has one zero mode .
the square with one diagonal brace is fragile because though it is rigid , if you cut / break any of the braces , it becomes floppy .
the square with two diagonal braces is firm because it remains rigid if you cut / break any one of the braces .
now consider the three magnetic structures , four spins on a square , three spins on a triangle and four spins on a tetrahedron in the figure .
in the square case , we subject the spins to @xmath310 for @xmath8 , @xmath311 nearest neighbor , @xmath312 indexes bonds and @xmath313 . in the triangle and tetrahedron
, we merely impose that the total spin of the object is zero .
the square is the analog of the firm mechanical structure : after imposing @xmath314 on three of its four sides , the fourth side is redundant . removing one bond would not destroy the nel order of its ground state .
the triangle is the example of the fragile mechanical structure .
all three constraints @xmath315 that define its ground state are needed to set the spins to be 120@xmath316 apart ( up to global spin rotations ) .
if we drop one of these three constraints , it gains a zero mode . finally , the tetrahedron is the example of the floppy mechanical structure
. we can choose @xmath317 and @xmath318 independently and still @xmath319 .
in this light , the new isostatic magnetic order that we can transport from the theory of phonons , is an order like the triangle case : if we were to drop any one term in the hamiltonian that imposes a constraint on the ground state , there is no redundancy and the system becomes floppy .
isostatic magnetism is fragile .
remarkably , since we have already the other two cases , `` floppy '' magnetism in the form of highly frustrated magnets and `` firm '' magnetism for most other magnetic systems , it may just be a matter of time to identify isostatic magnetism at least within some temperature range .
one group of materials that may contain this novel magnetism is the family of distorted kagome antiferromagnets .
an undistorted ideal kagome antiferromagnet has @xmath47 but many zero modes .
adding distortions could then lift the ground state degeneracy without changing @xmath110 since distortions do not change the topology of the lattice network .
since distortioned kagome lattices are much more common than ideal kagome lattices , there may already be a number of such systems that are isostatic magnets . applied to each neighboring spins as shown with black arrows ( equivalent to a uniform field on each spins ) . ]
another physical phenomena we can transport from the phonon theory is the existence of self field modes that are the analog of self stress modes .
the square with two diagonal beams of fig .
[ fig : isomagnetism ] has such a self stress mode because of its redundancy of constraints .
if one puts the diagonal beams in tension @xmath320 , then one can place the horizontal and vertical beams in tension @xmath321 and no force will be exerted on the pins at is vertices .
similarly , one can add a `` bond magnetic field '' @xmath322 via a term in the hamiltonian @xmath323 and there must exist @xmath322 such that no net field is experienced by a spin to linear order in the displacements of the spins from their ground states .
we know actually that there are at least @xmath324 of these from the topological index which here is @xmath325 . because we include spin rotation invariance , there are two zero modes and so @xmath326 demands @xmath327 . if @xmath3 labels bonds cyclically around the square , three of these six modes span @xmath328 .
these produce @xmath329 non - linearly ( i.e. independent of setting @xmath330 ) . of the remaining three ,
one is the case @xmath331 assuming we choose a neel ground state in the @xmath332-direction .
this case produces @xmath333 only after we linearize the spins @xmath334 around their magnetic ground state by setting @xmath330 . to non - linear order in the deviations of the spins from their ground state , this case will produce a non - zero energy @xmath335 .
the remaining two self field modes behave similarly .
so self field modes exist and extend our understanding of the theory of magnetism .
actually , we can understand what an experimentalist could achieve with a knowledge of self field modes in analogy with the self stress modes .
self stress modes have been studied at least as far back as eshelby s 1957 paper@xcite who emphasized the engineering importance of self stresses in an elastic medium with inhomogeneities .
more recently they have been understood as a means of selective mechanical failure of metamaterials@xcite . in a @xmath47 lattice with one self stress mode and one zero mode located in different regions ,
a load was shown to cause mechanical failure exactly at the self stress mode bonds .
similarly , an applied field that corresponds to a self field mode should cause a spin flip .
indeed , in the case above for the square with a field @xmath336 applied , we see that the spins will flip due to the non - zero energy @xmath337 caused by the non - linearities .
the other two self field modes not discussed in detail similarly cause a spin flip , just not all spins flip in those cases .
so such self field modes correspond to spin flip transitions and their identification in the constraint matrix eigenvalue problem may enable the discovery of a variety of such spin flip transitions .
in this paper , we have recognized a supersymmetry in the structure of hamiltonians that energetically impose local constraints on their ground states .
this allowed us to formally connect recent development in the theory of phonons to a wider range of systems including frustrated and unfrustrated magnets .
the connection ultimately arose from the similar role of the vanishing extension of springs @xmath15 of phonon ground states to the vanishing total spin on a bond in unfrustrated magnets and vanishing total spin on a triangle , square , tetrahedron or other local simplex in frustrated magnetic systems .
it also highlights the role of dirac s constraint matrix as a central pillar of these recent developments as well as proves that a topological index exists in all these systems .
lets end our discussion of these results with some final comments .
remarkably , this supersymmetry identifies maxwell s counting of degrees of freedom @xmath338 , where @xmath339 is the total number of degrees of freedom and @xmath340 is the total number of local constraints imposed by the hamiltonian ( i.e. possibly a redundant set ) , to the witten index @xmath110 associated with the supersymmetry .
this maxwell counting , though long used for balls and springs and other mechanical structures , was proposed by moessner and chalker as a means of characterizing highly frustrated antiferromagnets@xcite .
here we have connected their work with the actual number of zero modes through recognizing it as the topological index .
lattice where dark blue are the mn atoms with magnetic moments and yellow atoms are the ir atoms without magnetic moments .
one view of the lattice is through the 111 planes for the blue magnetic atoms .
they form a kagome lattice .
another view is achieved by considering how each triangle shares a side with another triangle in the octahedron .
this view suggests irmn@xmath2 is composed of side sharing triangles in 3 dimensions and is therefore more similar to the triangular lattice antiferromagnet than the kagome antiferromagnet . ]
the use of analogs of maxwell constraints in magnetism and the associated topological index provides a simple picture of the strength of magnetism in a system . to see this in practice ,
consider the question of the origin of the industrial strength of antiferromagnetism in irmn@xmath2 .
this material magnetically orders in a 120@xmath341 coplanar state at 960 k@xcite but is composed of kagome planes with a dominant isotropic heisenberg spin exchange@xcite that would appear to frustrated the magnetism@xcite .
how then can the maxwell - like constraints advanced by this paper explain such industrial strength magnetism ?
the answer lies in the many different kagome planes .
they all interconnect as shown in fig .
[ fig : irmn3 ] .
indeed , the lattice is better thought of as side - sharing triangles in three dimensions instead of layers of kagome planes . with eight triangles in the unit cell , the dominant heisenberg term in its hamiltonian can be written @xmath342 where @xmath343 is the total spin on each triangle .
we therefore have a topological index @xmath344 ( since there are 8/3 triangles per spin ) .
we can contrast this to say an unfrustrated antiferromagnet on the cubic lattice with three bonds in the unit cell and @xmath345 .
but upon closer inspection , we should be careful with such a comparison . in irmn@xmath2 ,
nearest neighbor heisenberg exchange still has a sub - extensive number of zero modes because the side sharing triangles form corner sharing octehedra .
so its @xmath346 is likely lower than the unfrustrated antiferromagnet with the same @xmath347 .
nevertheless , this effect is subextensive so it is unlikely to be a big effect .
therefore , the energetic constraints imposed by the dominant heisenberg exchange in irmn@xmath2 appear to be nearly as redundant as an unfrustrated antiferromagnet on a cubic lattice suggesting great industrial strength even though it is composed of kagome planes .
but not only firm magnetism may prove useful for practical applications .
fragile isostatic magnetism may prove to be easily manipulated for special design purposes .
it may play a role for magnetic systems similar to the role isostatic phonons have played in the design of metameterials . due to the absence of redundant constraints
, isostatic lattices can have particular weaknesses that can enable their manipulation .
perhaps even metamaterials designs that exploit this could carry over to magnetic materials .
for example , one lattice was designed to have selective mechanical failure in ref . where just a local set of bonds associated with a state of self stress broke down under a load .
the magnetic analog of this could correspond to a material design with a selective spin flip of a local subset of spins achieved with a global magnetic field .
in this way , isostatic magnetism could prove to have an important role in the design of magnetic materials for applications . and given packing fraction @xmath348 . here
the isostatic point highlighted by the green dot at the top right occurs for frictionless grains .
b ) the phase diagram for quantum disordering a frustrated antiferromagnet obtained in a semi - classical field theory@xcite .
the x - axis is a dimensionless temperature and the y - axis a coupling measuring the strength of quantum fluctuations .
c ) classical phase diagram of the loss of magnetic order based on the results of this paper . here again the green dot represents the isostatic point .
notice in all three cases , a special point exists in the phase diagram that sheds light on the rest of the phase diagram .
it would be interesting to understand the relationship between these three points.,title="fig : " ] and given packing fraction @xmath348 . here
the isostatic point highlighted by the green dot at the top right occurs for frictionless grains .
b ) the phase diagram for quantum disordering a frustrated antiferromagnet obtained in a semi - classical field theory@xcite .
the x - axis is a dimensionless temperature and the y - axis a coupling measuring the strength of quantum fluctuations .
c ) classical phase diagram of the loss of magnetic order based on the results of this paper . here
again the green dot represents the isostatic point .
notice in all three cases , a special point exists in the phase diagram that sheds light on the rest of the phase diagram .
it would be interesting to understand the relationship between these three points.,title="fig : " ] and given packing fraction @xmath348 . here
the isostatic point highlighted by the green dot at the top right occurs for frictionless grains .
b ) the phase diagram for quantum disordering a frustrated antiferromagnet obtained in a semi - classical field theory@xcite .
the x - axis is a dimensionless temperature and the y - axis a coupling measuring the strength of quantum fluctuations .
c ) classical phase diagram of the loss of magnetic order based on the results of this paper . here
again the green dot represents the isostatic point .
notice in all three cases , a special point exists in the phase diagram that sheds light on the rest of the phase diagram .
it would be interesting to understand the relationship between these three points.,title="fig : " ] but isostatic magnetism may also extend our theoretical understanding of magnetic materials .
there is likely a separate topological property not discussed in this paper .
since the magninos are free fermions , and the ideal isostatic magnetism state is a gapped state , it must fit into our categorization of topological band theory@xcite much like the phoninos where shown to have a topological band theory by kane and lubensky ( though they did nt call them phoninos ) .
it is the supersymmetry established by this paper that enables physics that applies to free fermions ( including majorana fermions ) to also apply to phonons and magnons .
this is one of the benefits of generalizing kane and lubensky s theory to a broader class of systems .
the results of this paper may also shed light on a long standing problem in the field of magnetism beyond band theory : how to melt magnetic order .
the connection established by our formal developments shows explicitly that the loss of rigidity in a solid@xcite , the jamming transition@xcite and the loss of magnetic order due to frustration[ramirez1994,moessner1998a , moessner1998b ] may all be different aspects of the same problem ( see fig .
[ fig : jamming ] ) .
each of these systems are built around constraints , have a topological index @xmath110 , and have isostatic order with @xmath47 that we curiously identify here with spontaneous susy breaking .
so from this perspective , isostatic magnetism is at a critical point between magnetic order and a classical spin liquid . it would be interesting to understand how this critical point is related to the quantum melting of antiferromagnetic order through a quantum critical point@xcite particularly given theories of this quantum phase transition that are semi - classical in nature .
* practical use in the design of magnetic materials via the topological index , zero modes and self field modes * a novel application of topological band theory of fermions to magnons * insight into the mechanism of quantum spin liquids and other exotic phases of magnetism .
47ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty http://dx.doi.org/10.1038/nphys2835
10.1038/nphys2835 http://www.nature.com/nphys/journal/v10/n1/abs/nphys2835.html{#}supplementary-information [ * * , ( ) ] link:\doibase 10.1016/0020 - 7683(78)90052 - 5 [ * * , ( ) ] link:\doibase 10.1088/1367 - 2630/15/4/043043 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.103.248101 [ * * , ( ) ] link:\doibase 10.1103/physrevx.5.031011 [ * * , ( ) ] link:\doibase 10.1038/nphoton.2014.248 [ * * , ( ) ] , link:\doibase 10.1038/nphys3228 [ * * , ( ) ] link:\doibase 10.1103/physrevb.87.144101 [ * * , ( ) ] link:\doibase 10.1103/physrevb.89.134409 [ * * , ( ) ] link:\doibase 10.1038/nphys3232 [ * * , ( ) ] http://arxiv.org/abs/1503.01324 [ , ( ) ] , link:\doibase 10.1073/pnas.1405969111 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.116.135501 [ * * , ( ) ] link:\doibase 10.1038/nature08917 [ * * , ( ) ] http://article.pubs.nrc-cnrc.gc.ca/ppv/rpviewdoc?issn=1208-6045{&}volume=79{&}issue=11-12{&}startpage=1283{&}ab=y file:///users / mlawler / dropbox / documents / papers/2001/moessner / canadian journal of physics 2001 moessner.pdf [ * * , ( ) ] http://projecteuclid.org/dpubs?service=ui{&}version=1.0{&}verb=display{&}handle=euclid.jdg/1214437492$\backslash$nhttp://www.math.toronto.edu/mgualt/morse theory / witten morse theory and supersymmetry.pdf [ * * , ( ) ] http://arxiv.org/abs/1407.2890 [ , ( ) ] , link:\doibase 10.1126/science.1248253 [ * * , ( ) ] @noop link:\doibase 10.1098/rspa.1958.0141 [ * * , ( ) ] http://link.aps.org/doi/10.1103/physrevlett.80.2929 papers://ab7c6739 - 3c9b-4bbe-8ce8 - 067dfd022c15/paper / p6318
file:///users / mlawler / dropbox / documents / papers/1998/moessner / physical review letters 1998 moessner.pdf [ * * , ( ) ] link:\doibase 10.1007/bf01325811 [ * * , ( ) ] link:\doibase 10.1007/bf02729860 [ * * , ( ) ] link:\doibase 10.1016/0003 - 4916(77)90335 - 9 [ * * , ( ) ] link:\doibase 10.1143/ptp.63.599 [ * * , ( ) ] link:\doibase 10.1016/0370 - 1573(94)00080-m [ * * , ( ) ] , @noop _ _ ( , ) link:\doibase 10.1103/physrevlett.116.135503 [ * * , ( ) ] link:\doibase 10.1103/physrevb.86.115131 [ * * , ( ) ] link:\doibase 10.1103/physrevb.86.125119 [ * * , ( ) ] link:\doibase 10.1103/physrevb.90.115141 [ * * , ( ) ] link:\doibase 10.1007/bf01197577 [ * * , ] @noop _ _ ( , ) link:\doibase 10.1103/physrevb.58.12049 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.99.137207 [ * * , ( ) ]
link:\doibase 10.1098/rspa.1957.0133 [ * * , ( ) ] link:\doibase 10.1038/nphys3185 [ * * , ( ) ] @noop * * ( ) link:\doibase 10.1103/physrevb.79.020403 [ * * , ( ) ] link:\doibase 10.1038/nature06981 [ * * , ( ) ] link:\doibase 10.1103/physrevb.39.2344 [ * * , ( ) ] @noop * * ( ) \doibase http://dx.doi.org/10.1016/s0022-5096(02)00107-2 [ * * , ( ) ] \doibase http://dx.doi.org/10.1016/j.jmps.2005.10.008 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.81.1841 [ * * , ( ) ] link:\doibase 10.1038/23819 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.61.1029 [ * * , ( ) ] | i generalize the theory of phonon topological band structures of isostatic lattices to frustrated antiferromagnets .
i achieve this with a discovery of a many - body supersymmetry ( susy ) in the phonon problem of balls and springs and its connection to local constraints satisfied by ground states .
the witten index of the susy model demands the maxwell - calladine index of mechanical structures .
`` spontaneous supersymmetry breaking '' is identified as the need to gap all modes in the bulk to create the topological isostatic lattice state .
since ground states of magnetic systems also satisfy local constraint conditions ( such as the vanishing of the total spin on a triangle ) i identify a similar susy structure for many common models of antiferromagnets including the square , triangluar , kagome , pyrochlore nearest neighbor antiferromagnets , and the @xmath0 square lattice antiferromagnet . remarkably , the kagome family of antiferromagnets is the analog of topological isostatic lattices among this collections of models .
thus , a solid state realization of the theory of phonon topological band structure may be found in frustrated magnetic materials .
recently , kane and lubensky@xcite ( kl ) identified topological properties of isostatic lattice phonons .
they achieved this by discovering a topological index governing mechanical structures by building on calladine s work@xcite and further utilizing a dirac - like square - root of the phonon equations of motion , a problem to which they could apply the theory of topological insulators .
remarkably , they showed the existence of lattices with gapped phonons for periodic boundary conditions that must have gapless phonons with open boundary conditions .
it is natural to wonder if this striking effect is more general and they conclude their study with `` finally , it will be interesting to explore connections with theories of frustrated magnetism@xmath1 . ''
where reference 48 is my study@xcite identifying topological gauge dynamics of the zero modes of classical kagome antiferromagnets .
this paper s goal is to make this connection and use it to generalize kl theory beyond phonons .
the kl theory of isostatic lattices is a different branch of topological phases from the theory of topological band insulators . following the original discovery of topological insulators , topological properties of boson band structures
have been studied for a wide variety of systems including phonons@xcite , photons@xcite , acoustic phononic crystals@xcite and magnons@xcite .
these systems achieve their topological properties in the presence of time reversal symmetry breaking and are built directly from the physics of the integer quantum hall effect .
in contrast , kl s theory of isostatic phonons is time reversal symmetric and `` purely geometric in nature''@xcite .
a connection to the integer quantum hall effect is made only after a dirac - like square - rooting procedure of the equations of motion .
it therefore presents a new direction in the theory of topological phases .
remarkably , though its may not apply directly to solid state phonons because they are mechanically stable , the kl theory has already seen a variety of applications due to its insight into the general phenomena of mechanical collapse .
these include some topological aspects of the jamming and rigidity percolation transitions@xcite , metamaterials made from beams and pins@xcite , and , remarkably , origami@xcite .
ref . has also taken it beyond the linearized level and discovered solitons that can propagate freely in the bulk of the isostatic state .
so , given the fundamental insight it provides , any extension of kl theory to a new class of systems , including extensions that go beyond the linearized limit , is likely to shed new light on those systems . in this light , frustrated magnets and/or
highly frustrated magnets are a prime target for an extension of kl theory .
they are magnets not only `` on the verge of collapse '' but also those that have already `` collapsed '' .
here collapsing is the analog of destabilizing the magnetic ordered state into a paramagnetic state such as a quantum spin liquid or valence bond solid .
a variety of materials including the organics , kagome family and pyrochlores are heavily studied for this reason@xcite .
in addition , highly geometrically frustrated magnets have a form of accidental degeneracy that results from a special feature of the spin hamiltoinan@xcite .
this frustration is toy - like ( fine tuned ) perhaps in a similar way that balls and springs are toy - like versions of a general theory of phonons .
so if kl theory were applicable to frustrated magnets , this generalization might apply to many already realized solid state materials . in this paper
, we show that the key to generalizing kl theory to other systems lies in connecting it to an abstract theoretical framework that takes the form of a many - body supersymmetric ( susy ) structure that extends the description of balls and springs .
the new fermionic degrees of freedom , that i dub `` phoninos '' , are superpartners to phonons and are governed by the kl theory s square - rooted equations of motion . for linearized phonons , the two sets of degrees of freedom are decoupled .
the phoninos therefore need not be real degrees of freedom but just reflect the specialness of balls and springs compared to a more general theory of phonons .
i then show that the topological index identified in kl theory is demanded by the witten index@xcite of susy .
a similar conclusion was reached by ref . in the continuum limit .
remarkably , spontaneous susy breaking here is just the need to gap all modes in the bulk to create the topological state ( as could be the case for topological superconductors@xcite ) . finally , we discuss what protects this topology including the role of quantum effects and non - linearities .
i then apply the same many - body susy construction to the case of magnons .
it turns out susy is found for both unfrustrated and frustrated magnons for either case has ground states that satisfy local constraints and the susy construction is built on these constraints .
remarkably , kagome magninos are governed by a hamiltonian which is the dirac s constraint matrix@xcite studied in ref . and is related through susy to the ordinary kagome magnon problem . the witten index in this case
is then shown to vanish for periodic boundary conditions but not for open boundary conditions demonstrating that kagome magnons are the analog of the isostatic lattice of kl theory if we can gap all their modes without changing topology such as in a distorted kagome crystal .
such a phase we call isostatic magnetism .
we conclude by identifying some of the new physical phenomena predicted by these formal development : the design of magnets by exploiting weaknesses in isostatic magnetism such as through a rich set of magnetic field induced spin flip transitions that we call `` self field modes '' ; and an application of the ideas here to the origin of the industrial strength of mn@xmath2ir ( why it is a robust antiferromagnet ) ; the potential application of topological band theory of electrons to magninos in the isostatic magnetic phase and via susy to magnons ; the study of the loss of magnetic order via isostatic magnetism which can be viewed as a classical critical point between magnetic order and a classical spin liquid@xcite / cooperative paramagnet@xcite . |
the possibility of quantum - gravity - induced in - vacuo dispersion , an energy dependence of the travel times of ultrarelativistic particles from a given source to a given detector , has been motivated in several studies ( see _ e.g. _ refs.@xcite and references therein ) .
part of the interest in this possibility comes from the fact that it is a rare example of candidate quantum - gravity effect that could lead to observably large manifestations , even if , as it appears to be safe to assume , its characteristic length scale is of the order of the minute planck length ( inverse of the planck energy scale @xmath0 ) or anyway not much larger than that .
the best opportunity so far studied for such experimental tests is provided by observations of grbs @xcite , which set up for us a sort of race among photons of different energies and ( probably @xcite ) neutrinos of different energies , all emitted within a relatively small time window . the fact that our understanding of the mechanisms producing grbs remains preliminary
is a challenge , since any given time - of - arrival difference among two particles can in principle always be attributed to the emission mechanism , but this can be compensated by suitable techniques of statistical analysis . for more than a decade
the analyses of grb data from the in - vacuo - dispersion perspective were done considering only photons and focusing on what could be tentatively inferred from each single grb .
recently , thanks mainly to the icecube telescope , it became possible to contemplate the possibility that we might be observing also some grb neutrinos affected by in - vacuo dispersion ; moreover , for grb photons the abundance of observations cumulatively obtained by the fermi telescope reached the level sufficient for attempting to perform statistical analyses over the whole collection of fermi - observed grbs . some of us were involved in the first studies using icecube data for searching for grb - neutrino in - vacuo - dispersion candidates @xcite .
intriguing statistical analyses of in - vacuo dispersion over the whole collection of fermi - observed grbs were perfomed in a series of studies by bo - qiang ma and collaborators @xcite .
the neutrino studies of refs .
@xcite led to exposing a feature in the icecube neutrino data which could plausibly be a manifestation of in - vacuo dispersion .
the possibility that this feature could be the result of background neutrinos just accidentally arranging themselves as if they were grb neutrinos affected by in vacuo dispersion was considered using statistical tools of analysis , finding that it would be very untypical " for background neutrinos to produce accidentally such a pronounced in - vacuo dispersion feature .
the grb - photon studies reported in refs .
@xcite also led to exposing a feature which could be a manifestation of in - vacuo dispersion . while this feature for grb photons is certainly striking , as observed most convincingly in ref .
@xcite , there was so far no attempt to characterize quantitatively its statistical significance .
the main objective of the study we are here reporting is to characterize the statistical significance of the feature exposed in refs .
@xcite for photons , and to show that this feature is surprisingly consistent with the feature exposed in refs .
@xcite for neutrinos of much higher energies ( the relevant photons have energies of the order of 10 gev , while the neutrinos have energies of the order of 100 tev ) .
we also offer some preliminary observations which might become relevant if any of the features here contemplated find greater support as more data are accrued , concerning the possible interpretation of such features as manifestations of ( so far unknown ) astrophysical mechanisms , rather than as manifestations of in - vacuo dispersion .
the class of scenarios we intend to contemplate finds motivation in some much - studied models of spacetime quantization ( see , _
e.g. _ , @xcite and references therein ) and , for the type of data analyses we are interested in , has the implication that the time needed for a ultrarelativistic particle to travel from a given source to a given detector receives a quantum - spacetime correction , here denoted with @xmath1 .
we focus on the class of scenarios whose predictions for energy ( @xmath2 ) dependence of @xmath1 can all be described in terms of the formula ( working in units with the speed - of - light scale @xmath3 " set to 1 ) @xmath4 here the redshift- ( @xmath5-)dependent @xmath6 carries the information on the distance between source and detector , and it factors in the interplay between quantum - spacetime effects and the curvature of spacetime . as usually done in the relevant literature @xcite we take for @xmath6 the following form : ) is by far the most studied scenario , some alternatives to ( [ dz ] ) are also under consideration @xcite . ]
@xmath7 where @xmath8 , @xmath9 and @xmath10 denote , as usual , respectively the cosmological constant , the hubble parameter and the matter fraction , for which we take the values given in ref.@xcite .
with @xmath11 we denote the planck scale ( @xmath12 ) while the values of the parameters @xmath13 and @xmath14 in ( [ main ] ) characterize the specific scenario one intends to study . in particular ,
in ( [ main ] ) we used the notation @xmath15 " to reflect the fact that @xmath14 parametrizes the size of quantum - uncertainty ( fuzziness ) effects .
instead the parameter @xmath13 characterizes systematic effects : for example in our conventions for positive @xmath13 and @xmath16 a high - energy particle is detected systematically after a low - energy particle ( if the two particles are emitted simultaneously ) .
the dimensionless parameters @xmath13 and @xmath14 can take different values for different types of particles @xcite , and it is of particular interest for our study that in particular for neutrinos some arguments have led to the expectation of an helicity dependence of the effects ( see , _ e.g. _ , refs.@xcite and references therein ) .
therefore even when focusing only on neutrinos one should contemplate four parameters , @xmath17 , @xmath18 , @xmath19 , @xmath20 ( with the indices @xmath21 and @xmath22 referring of course to the helicity ) .
analogous considerations apply to photons and their polarization @xcite .
the parameters @xmath23 are to be determined experimentally . when non - vanishing , they are expected to take values somewhere in a neighborhood of 1 , but values as large as @xmath24 are plausible if the solution to the quantum - gravity problem is somehow connected , as some arguments suggest @xcite , with the unification of non - gravitational forces , while values smaller than 1 find support in some renormalization - group arguments ( see , _
e.g. _ , ref.@xcite ) . following refs .
@xcite , we find convenient to introduce a distance - rescaled energy " @xmath25 defined as and a distance - rescaled energy " @xmath25 , in refs .
@xcite , which focused on neutrinos , the analysis was set up in terms of correlations between energy and a distance - rescaled time delay " @xmath26 .
the two setups are evidently equivalent , but the one we adopt here is best suited for handling the possibility of a ( roughly-)systematic time offset at the source ( see later ) . for the values of @xmath1 that are relevant for the neutrino part of the analysis this possibility of a time offset
has a negligible role , and therefore the two setups are actually equally convenient , but for part of the analysis based on photons it is advantageous to set up the analysis in terms of correlations between @xmath1 and a distance - rescaled energy " @xmath25 . ]
@xmath27 so that ( [ main ] ) can be rewritten as @xmath28 this reformulation of ( [ main ] ) allows to describe the relevant quantum - spacetime effects , which in general depend both on redshift and energy , as effects that depend exclusively on energy , through the simple expedient of focusing on the relationship between @xmath1 and energy when the redshift has a certain chosen value , which in particular we chose to be @xmath29 . if one measures a certain @xmath1 and the redshift @xmath5 of the relevant grb is well known , then one gets a firm determination of @xmath25 by simply rescaling the measured @xmath2 by the factor @xmath30 . and
even when the redshift of the relevant grb is not known accurately one will be able to convert a measured @xmath2 into a determined @xmath25 with accuracy governed by how much one is able to still assume about the redshift of the relevant grb . in particular ,
even just the information on whether a grb is long or short can be converted into at least a very rough estimate of redshift .
eq.([maintwo ] ) , which follows the strategy of analysis proposed in refs .
@xcite , is ideally structured to handle the possibility that there be a ( roughly ) systematic time offset at emission between the time of emission of the low - energy particles used as reference ( we shall later take as reference the time of observation of the first peak of the low - energy - gamma - ray component of a grb ) and the higher - energy particle of interest .
such an astrophysical mechanism for time offset at the source , would imply , within the modelization we are assuming for the quantum - spacetime effects , that @xmath1 is not exactly proportional to @xmath25 , since the observed @xmath1 would receive both a contribution from the quantum - spacetime effects given by the right - hand side of eq.([maintwo ] ) and a contribution due to the time offset at the source .
this latter contribution can be described as @xmath31 , where @xmath32 is the time offset at the source and the factor @xmath33 takes into account time dilatation . following refs .
@xcite these observations can be fruitfully used to replace eq.([maintwo ] ) with @xmath34 notice that in allowing for the mentioned possibility of a time offset at the source we also found appropriate to set up our equation as a relationship between @xmath35 and @xmath36 , so that the term involving @xmath32 is just a constant contribution , redshift independent and energy independent .
later , in our graphs showing @xmath35 versus @xmath36 , this will facilitate the visualization of @xmath32 .
we stress that here , just like in refs .
@xcite , we shall not allow for different values of @xmath32 for different photons for different photons one could never test the in - vacuo - dispersion hypothesis , since any measured value of @xmath1 could always be attributed to a corresponding value of @xmath32 at the source . ] .
we just allow for one value of @xmath32 valid for all photons of all grbs in the analysis , and we shall show that the present data situation fits rather nicely this apparently simplistic assumption .
as stressed above , the main objective of the study we are here reporting is to characterize the statistical significance of the in - vacuo - dispersion feature exposed in refs .
@xcite for photons , and to show that this feature is surprisingly consistent with the feature exposed in refs .
@xcite for neutrinos .
a quantitative characterization of the statistical significance of the in - vacuo - dispersion feature found for neutrinos was already given in ref .
@xcite , so we shall not have new results about that here , but it is still valuable for our purposes to summarize the main steps of that analysis .
this will also give us a chance to arrange the presentation in terms of correlations between @xmath37 and @xmath38 , whereas in ref .
@xcite the analysis was arranged in terms of correlations between energy and a distance - rescaled time - of - arrival difference " .
these two arrangements of the analysis are evidently equivalent for the neutrino case and @xmath38 . ] , but it is a good preparation for the later discussion of the photon case to have the discussion of neutrinos arranged in terms of correlations between @xmath37 and @xmath38 . for the neutrino case a crucial role is played by the criteria used for selecting some grb - neutrino candidates .
this is not at all an easy task since the present situation is such that at best we can catch a single neutrino from a whole grb .
moreover , in testing the hypothesis of in - vacuo dispersion , we must allow for a sizable time - of - observation difference between the neutrino and the first peak in fermi s gbm .
in fact , at the scales of interest for this neutrino analysis , involving in particular neutrinos with energy of order 100 tev , in - vacuo dispersion could produce values of @xmath1 of anything between a few hours and a couple of days .
of course such criteria for selecting grb - neutrino candidates will involve a temporal window ( how large can the @xmath1 be in order for us to consider a icecube event as a potential grb - neutrino candidate ) and some criteria of directional selection ( how well the directions estimated for the icecube event and for the grb should agree in order for us to consider that icecube event as a potential grb - neutrino candidate ) . a previous more preliminary analysis ( based on icecube data from june 2008 to may 2010 )
had tentatively put in focus a range of values of @xmath39 ( for our eq.([main ] ) somewhere in the range between 10 and 20 , and this we used in refs .
@xcite ( based on icecube data from june 2010 to may 2014 ) to choose a temporal window large enough to accommodate the corresponding size of the effects .
we took a temporal window of 3 days , and focused on icecube events with energy between 60 tev and 500 tev .
widening the range of energies up to , say , 1000 tev would have imposed us a temporal window of about 6 days , rendering even more severe one of the key challenges for this sort of analysis , which is the one of multiple grb candidate partners for a single icecube event . as directional criteria for the selection of grb - neutrino candidates we considered the signal direction pdf depending on the space angle difference between grb and neutrino : @xmath40 , a two dimensional circular gaussian whose standard deviation is @xmath41 , asking the pair composed by the neutrino and the grb to be at angular distance compatible within a 2@xmath42 region . a key observation for our analysis ( based on the corresponding observation made in our ref .
@xcite ) is that whenever @xmath13 and/or @xmath14 do not vanish one should expect on the basis of ( [ mainnew ] ) a correlation between @xmath43 and @xmath38 . our data set
@xcite is for four years of operation of icecube @xcite , from june 2010 to may 2014 .
since the determination of the energy of the neutrino plays such a crucial role in our analysis we include only icecube shower events " ( for track events " the reconstruction of the neutrino energy is far more problematic and less reliable @xcite ) .
we have 21 such events within our 60 - 500 tev energy window , and we find that 9 of them fit our requirements for candidate grb neutrinos .
the properties of these 9 candidates that are most relevant for our analysis are summarized in table 1 and figure 1 . ) , while black points correspond to early neutrinos " ( @xmath44 ) . ] as visible in table 1 , for some icecube events our selection criteria produce multiple grb - neutrino candidates . in ref.@xcite
we handled this issue of multiple candidates by focusing on the case that provides the highest correlation .
another issue reflected by table 1 comes from the fact that for only 3 of the grbs involved in this analysis the redshift is known .
we must handle only one short grb of unknown redshift , and we assume for it a redshift of 0.6 , which is a rather reasonable rough estimate for a short grb .
our 9 grb - neutrino candidates marked by an asterisk in table 1 include 8 long grbs , 2 of which have known redshift , and we assign to the other 6 long grbs the average @xmath45 of those two values of redshift ( @xmath46 ) . in figure 1
it is striking that the correlation between @xmath43 and @xmath38 gets stronger at higher energies .
interestingly , as observed in our ref.@xcite , this too fits the expectations of some quantum - spacetime models : as stressed in particular in ref.@xcite , in some of these quantum - spacetime models neutrinos can undergo processes of neutrino splitting " , and in turn this could plausibly @xcite affect a in - vacuo - dispersion study such as ours just in the way of rendering the correlation weaker at lower energies .
while this was worth mentioning , we shall here prudently not take it into account : we shall ignore neutrino splitting and handle all our 9 grb - neutrino candidates on the same footing .
the correlation between @xmath43 and @xmath38 for the 9 grb - neutrino candidates highlighted in fig.1 is of and @xmath38 .
these two types of correlation studies are based on two equations which are equivalent to each other , one obtained from the other by simply dividing both members of the equation by the same function of redshift . therefore in the ideal case of an infinite amount of data the indications emerging from the two types of correlation studies would be exactly coincident , but only 9 data points intervene in our analysis , spread over a wide range of values of redshift , and this results in the ( however small ) difference between 0.951 and 0.866 . ]
this is a strikingly high value of correlation but in itself does not provide what is evidently the most interesting quantity here of interest , which must be some sort of false alarm probability " : how likely it would be to have accidentally data with such good agreement with the expectations of the quantum - spacetime models here contemplated ?
we proposed in ref.@xcite that for these purposes one could estimate how often a sample composed exclusively of background neutrinos would produce accidentally 9 or more grb - neutrino candidates with correlation comparable to ( or greater than ) those we found in data .
we did this by performing @xmath47 randomizations of the times of detection of the 21 icecube neutrinos relevant for our analysis , keeping their energies and directions fixed , and for each of these time randomizations we redo the analysis just as if they were real data .
our observable is a time - energy correlation and by randomizing the times we get a robust estimate of how easy ( or how hard ) it is for a sample composed exclusively of background neutrinos to produce accidentally a certain correlation result .
also in the analysis of these fictitious data obtained by randomizing the detection times of the neutrinos we handle cases with neutrinos for which there is more than one possible grb partner by maximizing the correlation , in the sense already discussed above for the true data .
we ask how often this time - randomization procedure produces 9 or more grb - neutrino candidates with correlation @xmath48 , and remarkably we find that this happens only in 0.11@xmath49 of cases .
having correlation as high as 0.866 ( and false alarm probability of 0.11@xmath49 ) is particularly striking considering that surely at least some of our 9 grb - neutrino candidates are just background neutrinos accidentally fitting our criteria for the selection of grb - neutrino candidates .
this can be straightforwardly deduced by observing that out of the 21 neutrinos in our sample , since indeed only 9 turned out to fit our requirements for grb - neutrino candidates , there are at least 12 neutrinos which are background ) is correct they are to be considered as background since they do not fit our selection criteria , and of course those 12 neutrinos are background also if eq.([main ] ) is not correct ( in that case all the 21 events are not grb neutrinos ) . ] .
we can therefore ask how likely it would have been for one or more of those 12 neutrinos to accidentally appear to be grb neutrinos of the type we are looking for .
this can be estimated by randomizing the times of those 12 neutrinos .
of course , if , say , it is likely that 3 of those neutrinos could appear as grb neutrinos , we will assume that a proportionate number of our 9 grb - neutrino candidates are background .
details on this line of reasoning are given in ref .
important for us here is that this line of reasoning leads to the conclusion that it is very likely that at least 3 of our 9 grb - neutrino candidates are background .
this renders somewhat striking the fact that , in spite of these contributions by background neutrinos , we found a value of correlation as high as 0.866 . concerning our notion of false alarm probability " this deduction about
the role played by background neutrinos may suggest a somewhat different strategy of analysis @xcite .
let us illustrate this by taking as working assumption that 3 among our 9 grb - neutrino candidates surely are background .
we can then exploratively assume that the 6-plet of true " grb neutrinos is the maximum - correlation 6-plet among the 6-plets obtainable from our 9 candidates , and take as reference for the analysis the value of correlation found for this maximum - correlation 6-plet , which is 0.995 .
one can then define a false alarm probability based on how frequently simulated data , obtained by randomizing the times of detection of all the 21 neutrinos in our sample , include a 6-plet of candidates with correlation greater or equal to the value of 0.995 found for the best 6-plet in the real data ( so , if , say , a given time randomization produces 11 candidates one would assign to the randomization a value of correlation given by the highest correlation found by considering all possible choices of 6 out of the 11 candidates ) .
we find that this false alarm probability is of 0.6@xmath49 .
having reviewed briefly the case for in - vacuo dispersion for neutrinos " , and the characterization of its statistical significance provided in our ref .
@xcite , we are ready to proceed with our analysis of the case for in - vacuo dispersion for photons " , emerging from the investigations reported in refs . @xcite . for this photon case , while ref .
@xcite convincingly characterized the relevant feature as striking , there was so far no characterization of the statistical significance , so one of our main objectives here is to provide such a characterization . the analyses reported by ma and collaborators in refs .
@xcite focus on the highest - energy photons among those observed for grbs by the fermi telescope , and implements some time - window selection criteria .
evidently , in spite of the many differences between the two contexts , there are challenges in this sort of analysis of grb photons , which are rather similar to the challenges faced in the analysis of candidate grb neutrinos reported above .
we find appropriate to here contemplate not only the energy - window and time - window criteria adopted by ma and collaborators but also to propose some alternative criteria of our own , which ( while keeping close to the criteria introduced by ma and collaborators ) we feel might be a natural alternative to be considered as this research program further advances , especially as new data are accumulated .
ma and collaborators focus on grb photons observed within 90 seconds of the first peak in the gbm and with observed energy greater than 10 gev . in our alternative criteria
we choose to specify the time window by mainly exploiting the fact that , as already observed in ref .
@xcite , a surprisingly high percentage of the photons selected by the criteria of ma and collaborators are consistent with roughly the same value of the time offset at the source @xmath32 .
we attempt to exploit this aspect in our time - window selection criteria by essentially characterizing the time window in terms of emission times , rather than observed times .
we require that at the source the time of emission of our selected photons be consistent with an offset with respect to the time of emission of the first gbm peak of up to 20 seconds , but of course also allowing in addition for a sizeable range of effects possibly due to in - vacuo dispersion . when expressed in terms of the difference @xmath1 between the time of observation of the relevant photon and the time of observation of the first gbm peak , our time selection criterion takes the form @xmath50 here the @xmath51 are our mentioned window on @xmath32 , while the parameter we fix at @xmath52 allows for in - vacuo - dispersion effects of amount roughly comparable to the corresponding range of effects probed by ma and collaborators .
the main difference here is that our time window has the same quantitative interpretation for all grbs when described in terms of emission times at the source , but when expressed as a window on observed times it depends on the redshift of the grb .
the 90 seconds of redshift - independent observed - time window adopted by ma and collaborators roughly coincide with our window on observed times at redshift of 1 . for grbs at redshift greater than 1 ( where both time dilatation of the offset and the possible in - vacuo dispersion would produce bigger effects on the time of observation ) our eq.([timeselect ] ) allows for an observed - time window larger than 90 seconds , while for grbs at redshift smaller than 1 it allows for an observed - time window smaller than 90 seconds . for what concerns our window on photon energies ,
consistently with our focus on properties at the source ( rather than observed properties ) , we require that our selected photons be emitted at the source with energy greater than 40 gev .
so in terms of observed energy our window is @xmath53 , an alternative to the 10-gev redshift - independent observed - energy window of ma and collaborators .
we picked 40 gev as our cut on the energy at the source because the selection process for this choice gives results rather close to those obtained with the cut at 10 gev of observed energy adopted by ma and collaborators . at the present time
( as confirmed by our analysis ) there is no evidence that our criteria might be more advantageous than those of ma and collaborators .
we are only proposing them as an alternative which might play a role as this research program advances . accordingly , while we keep at center stage our proposed criteria , in this manuscript we shall also report some results that we obtained using the selection criteria of ma and collaborators .
an important final remark on selection criteria concerns redshifts . for our neutrino analysis it was possible , as shown above , to allow for grb - neutrino candidates for which the grb redshift had not been measured .
we expect , as argued more extensively in ref.@xcite , that by using as reference some estimated average value of redshifts for long and short grbs observed in neutrinos we should eventually , as more data is accrued , reach conclusive findings , in spite of handling grbs which , for the most part , have no precise redshift assignment .
such conclusive findings would have been reached faster in presence of more measured values of redshift , but without such measured values the analysis still works in the long run .
we believe , however , that for the analogous photon analyses the role of redshift measurements must be handled differently .
a challenge for this sort of photon analyses is that the size of the conjectured effects , often of a few seconds , is comparable to the time scales of the astrophysical mechanisms at work in a grb .
any eventual in - vacuo dispersion effect would have to be deduced finely within the sort of background noise " produced by the ( largely unknown ) mechanisms that cause the specific time variability of a given grb . as a result
we propose that in - vacuo - dispersion photon analyses should confine themselves to grbs of measured redshift .
also ma and collaborators rather strictly adopt this attitude toward redshifts , though they have handled cases where the grb redshift had been guessed on the basis of some theoretical argument but had not been measured .
we shall assume that it is safer for photon analyses to focus strictly on grbs on measured redshift .
we show in table 2 and figure 2 the 11 fermi - telescope photons selected by the time window of our eq.([timeselect ] ) and our requirement of an energy of at least 40 gev at emission .
the fact that our criteria are to a large extent compatible with the criteria of ma and collaborators is also suggested by figure 2 : all our 11 photons were also selected by ma and collaborators ; the only difference is that 2 of the photons selected by ma and collaborators are not picked up by our criteria .
these 2 additional photons are also shown in figure 2 and table 2 .
the content of figure 2 , as already efficaciously stressed in ref .
@xcite , is rather striking .
following ma and collaborators , we notice that all 13 photons ( the 11 picked up by our criteria , plus the two additional ones picked up by the criteria of ma and collaborators ) are well consistent with the same value of @xmath54 , upon allowing for only 3 values of @xmath32 .
we shall not however attempt to quantify the statistical significance of this more complex thesis based on 3 values of @xmath32 : evidently the most striking feature is that 8 of our 11 photons ( 9 of the 13 photons of ma and collaborators ) are all compatible with the same value of @xmath54 and @xmath32 .
this sets up a rather easy question that one can investigate statistically : if there is no in - vacuo dispersion , and therefore the correlation shown by the data is just accidental , how likely it would be for such 11 photons to include 8 that line up so nicely ? and @xmath38 .
gray points are for the 12th and 13th entries in table 2 ( photons picked up by the criteria of ma and collabroators , but not by our criteria ) . the strikingly visible feature of 8 black points ( plus 1 gray point ) falling nicely on a straight line
is also highlighted in figure by the presence of a best - fit line ( which however we find appropriate to discuss in detail only later , in section v ) . ]
we address this question quantitatively by first computing the correlation of the 8 among our 11 photons that line up nicely in figure 2 , finding that this correlation is 0.9959 .
we then estimate an associated false alarm probability " @xcite by performing simulations in which ( while keeping their energy fixed at the observed value ) we randomize , within the time window specified by our time - selection criterion , the time delay of each of our 11 high - energy photons with respect to the gbm peak of the relevant grb , and we assign to each of these randomizations a value of correlation given by the maximum value of correlation found by taking in all possible ways 8 out of the 11 photons .
we find that these simulated values of correlation are @xmath55 only in 0.0013@xmath49 of cases , about 1 chance in 100000 .
we stress that this impressive quantification of the statistical significance of the feature exposed by ma and collaborators does not depend on the fact that we adopted our own novel selection criteria . for this purpose
we redo the analysis including also the 2 photons that should be included according to the criteria of ma and collaborators .
in this case we have 9 out of 13 photons that line up very nicely .
the value of correlation found for those 9 photons is 0.9961 .
we then perform simulations in which we randomize the time of observation of all the 13 photons within the time window specified by the time - selection criterion of ma and collaborators and to each of these randomizations we assign a value of correlation given by the maximum value of correlation found by taking in all possible ways 9 out of the 13 photons .
we find that these simulated values of correlation are @xmath56 only in 0.0009@xmath49 of cases , very close to the 0.0013@xmath49 obtained with our selection criteria .
the values of correlation reported in the previous subsection , and especially the values of false alarm probability found in the previous subsection , are rather impressive . however , as discussed in the next section , the interpretation of these data presents us with some challenges . in light of this
we find appropriate to stress that the picture emerging from this photon feature has intrinsic model - independent predictive power " .
we illustrate this notion by considering the situation set up by the first two papers by ma and collaborators , refs .
@xcite , which were written before may 2016 ( _ i.e. _ before the observation of grb160509a ) . at that point ma and collaborators
had already discussed the photon feature using all the photons in our figure 2 , of course with the exception of the photon from grb160509a which had not yet been observed . that photon from grb160509a allowed then ma and xu , in ref .
@xcite , to appropriately emphasize that the picture was finding additional support . in a sense which we shall attempt to quantify ,
the picture ma and collaborators had been developing exhibited some predictive power upon the observation of grb160509a .
our quantification of this predictive power takes off by computing the value of correlation obtained with the other 8 photons on the main line " of figure 2 ( _ i.e. _ not including the photon from grb160509a , but including the photon on the main line " picked up by the selection criteria of ma and collaborators but not picked up by our selection criteria ) , finding that this correlation is of 0.9935 . with the observation of the photon from grb160509a the resulting 9-photon correlation moved up to 0.9961
we shall characterize the predictive power by asking how likely it would be for a photon unrelated to those previous 8 photons on the main line " to produce accidentally such an increase of correlation .
we randomize the time of observation of that photon from grb160509a ( within the time window specified by the time - selection criterion of ma and collaborators ) and we find that an increase of correlation from 0.9935 to 0.9961 ( or higher ) occurs only in 1.9@xmath49 of cases .
we perform the same estimate also adopting our selection criteria , as a mere academic exercise ( our selection criteria are being proposed here , after the observation of grb160509a ) . with our selection criteria
one has only 7 photons on the main line " , when considering data available before grb160509a .
those 7 photons have correlation of 0.9932 . adding
the photon from grb160509a one then has a 8-photon correlation of 0.9960 .
we randomize the time of observation of that photon from grb160509a ( within the time window specified by our time - selection criterion ) and we find that an increase of correlation from 0.9932 to 0.9960 ( or higher ) occurs only in 0.79 @xmath49 of cases .
our quantification of statistical significance gave rather impressive results both for the neutrino feature and for the photon feature .
we still feel that the overall situation should be assessed prudently , since both analyses still rely on only a small group of photons and neutrinos .
there is no reason to jump to any conclusions , also because both the fermi telescope and the icecube observatory will continue to report new data still for some time to come .
it is nonetheless interesting to assess the present situation both from the viewpoint of possible interpretations and from the viewpoint of a possible consistency between different analyses .
a first step of interpretation must concerns the 3 photons that in figure 2 do not line up with the other 8 photons , the 8 photons which lie on the main line " @xcite of ma and collaborators .
the tentative interpretation one must give within the setup of these analyses is that those 3 photons were not emitted in coincidence with the fist peak of the gbm signal .
the time window of our selection criterion ( and similarly the one of the selection criterion adopted by ma and collaborators ) is structured in such a way to
catch " those high - energy photons that were emitted roughly at the same time when the first peak of the gbm was emitted , but if truly in - vacuo dispersion is at work evidently it would happen occasionally that just because of in - vacuo dispersion some photons not emitted in coincidence with the first peak of the gbm end up being observed within our time window . while it is evidently difficult to quantify how frequently this should occur ,
at least qualitatively what is shown in figure 2 is just what one should expect if in - vacuo dispersion truly occurs , including the presence of some photons outside the main line " . as an aside , let us however notice that the significance of what is shown in figure 2 is not washed away if we include in the analysis of statistical significance also the photons outside the main line " .
for this purpose we first notice that the value of correlation obtained by taking into account all 11 photons is 0.845 , still rather high . in our simulations ,
in which we randomize the time of observation of the 11 photons ( within the time window specified by our time - selection criterion ) , we find that a value of correlation for all 11 photons @xmath57 0.845 is obtained only in 0.035 @xmath49 of cases .
similar conclusions are reached adopting the criteria of ma and collaborators . in that case one
has 13 photons under consideration , and the value of correlation computed for those 13 photons is 0.805 . randomizing the times of observation of those 13 photons ( within the time window specified by the time - selection criterion of ma and collaborators ) we find that a value of correlation for all 13 photons @xmath57 0.805 is obtained only in 0.037@xmath49 of cases . in light of the observations
made in the previous subsection one feels encouraged to set aside the 3 photons that fall off the main line " , and focus on the other 8 photons .
a significant characterization of those 8 photons is obtained by assuming @xmath58 , so that the whole feature is due to a nonzero value for @xmath59 .
this assumption @xmath58 is very restrictive but still the main line " of 8 photons in figure 1 is very well described by the model of eq.([mainnew ] ) , for @xmath60 and @xmath61 .
these are the parameters of the line shown in figure 2 , where the goodness of the fit of the 8 photons on the main line " is visible . the story with the main line " , the single time offset shared by 8 photons , and @xmath58 fits indeed in remarkably nice way , in spite of being based on very restrictive assumptions .
this is surely striking , but we are nonetheless inclined to proceed cautiously .
there is evidently a pronounced feature , of the type here characterized , in these available grb - photon data , but its description does not necessarily have to be the one that presently fits the data so nicely . first we should stress that in spite of our impressive findings for the false alarm probabilities , we still consider as most likely the hypothesis that the feature is accidental , rather than a truly physical ( in - vacuo - dispersion - like ) feature .
perhaps more surprisingly , even when taking as temporary working assumption that the feature is physical we give priority to the hypothesis that the feature might not really be describable in terms of the ingredients composed by the main line " ,
the same time offset for 8 photons and @xmath58 .
we are inclined to adopt this attitude because our intuitive assessment is that , even if the overall feature is physical , at least part of present picture , with these few data available , could be accidental .
we feel that such level of prudence is methodologically correct in general , and in this case might find additional motivation in the fact that the offset time favored by the analysis summarized in figure 2 would require , as observed above , a majority of our photons to have been emitted at the source some 11 seconds before the time of emission of the gbm peak .
( we might have had a slightly different intuition had we found a similar result but for 11 seconds after the gbm peak . )
we give tangibility to these considerations by taking temporarily as working assumption , as an illustrative example , a hypothesis such that the feature is truly physical but the way it manifested itself so far is in part accidental .
for this purpose we scramble " the nice picture of figure 2 by not taking under consideration the @xmath1 , time difference with respect to time of observation of gbm peak , but rather a @xmath62 , time difference with respect to trigger time of the gbm signal .
we do this just to probe the dependence of our results on the perspective adopted in the analysis : we would not really expect that @xmath62 is better than @xmath1 at exposing the sought correlation , but it is interesting for us to see whether the feature completely disappears by replacing @xmath1 with @xmath62 . with @xmath62 . ]
what we get upon relying on @xmath62 in place of @xmath1 is the picture given by figure 3 . in figure 3
there is no neat main line " , but this is after all what we would have expected before looking at the data : we would have expected the time offset at the source ( with respect to the first gbm peak or to the gbm trigger ) to be at least a bit different for different photons ; moreover , with a nonvanishing @xmath63 the time of observation of each photon would receive an additional random component .
importantly for our purposes , one should notice that , while figure 3 is surely less striking than figure 2 , the feature has not disappeared : it is less pronounced but the overall picture of figure 3 still shows a surprisingly high correlation .
this is what we mean by contemplating the hypothesis that only part of what is shown in figure 2 might be physical , with the rest being just accidental result of how these first 11 photons usable for our purposes happened to match very neatly a particular set of hypothesis for the interpretation and the analysis .
quantitatively we have that for the data analyzed in the way reflected by figure 3 we have correlation of 0.775 , over all 11 photons picked up by our selection criteria .
randomizing , within the time window specified by our time - selection criterion , the time delay of each of our 11 high - energy photons with respect to the gbm trigger of the relevant grb , we find correlation @xmath64 in only @xmath65 of cases , which is ( not as small as the @xmath66 found above for the analysis with @xmath1 , but ) still a very small false alarm probability . in light of the observations made in the previous subsection we now set aside the 3 photons that fall off the main line " , and focus on the other 8 photons .
a significant characterization of those 8 photons is obtained by assuming @xmath58 , so that the whole feature is due to a nonzero value for @xmath59 .
this assumption @xmath58 is restrictive but still the main line " of 8 photons in figure 1 is very well described by the model of eq.([mainnew ] ) , for @xmath67 and @xmath61 .
it is interesting to compare this estimate of @xmath59 with the estimate of @xmath39 that one can obtain from the neutrino data here discussed in section iii .
this comparison should be handled with some care , since some quantum - spacetime models predict ( see , _
e.g. _ , ref .
@xcite and references therein ) independent in - vacuo dispersion parameters for different particles , and also a possible dependence of the effects on polarization for photons and on helicity for neutrinos .
still one would tentatively expect comparable magnitude of the effects for different particles ( including the possible dependence on polarization / helicity ) .
a first important observation is the figure 1 includes @xcite 5 neutrinos whose interpretation in terms of in - vacuo dispersion would require positive @xmath39 and 4 neutrinos whose interpretation in terms of in - vacuo dispersion would require negative @xmath39 ( this is why in figure 1 we consider @xcite the absolute value of @xmath1 ) .
another complication for our purposes originates in the fact that , as mentioned , we have reasons @xcite to expect that 3 or 4 of those 9 grb - neutrino candidates are actually background neutrinos that happened to fit accidentally our profile of a grb - neutrino candidate .
what we can do is to attempt an estimate of the absolute value @xmath68 and to perform this estimate by assuming that 3 of the 9 grb - neutrino candidates are background : essentially we estimate @xmath68 for each possible group of 6 neutrinos among our 9 grb - neutrino candidates , and we combine these estimates into a single overall estimate .
this leads to the estimate @xmath69 .
so we have an estimate of @xmath61 and an estimate of @xmath69 , which are closely comparable , as theoretical prejudice would lead us to expect . perhaps more importantly , the hypothesis that both features are accidental should also face the challenge introduced by this correspondence of values .
if actually there is no in - vacuo dispersion both features should be just accidental .
all 9 of our grb - neutrino candidates would just be background neutrinos who happened to fit our criteria for selection of grb - neutrino candidates and whose energies and times of observation just happened to produce the high correlation shown in figure 1 . and similarly all 11 of the photons selected by our criteria would have accidentally produced the correlation visible in figure 2 : they would be photons whose time of observation ( with respect to the time of observation of the gbm peak ) is not really correlated with energy , the correlation with energy emerging just accidentally .
all these assumptions about neutrinos and photons are needed if there is no in - vacuo dispersion , with the additional observation that all these accidental facts end up producing comparable estimates of @xmath59 and @xmath68 .
the level of consistency " ( in the sense discussed above ) between the neutrino feature and the photon feature is visually illustrated in our figure 4 , showing both our 11 photons and our 9 grb - neutrino candidates in a plot of @xmath25 versus the absolute value of @xmath1 . and @xmath25 .
] so far we only considered two alternative hypotheses : either the two features shown in figures 1 and 2 are due to in - vacuo dispersion or there is no in - vacuo dispersion and those two features are accidental .
one should of course contemplate a third possibility : there might be no in - vacuo dispersion but still at least one of those two features is not accidental , but rather the result of some other physical mechanism .
this leads us naturally to wonder whether the two features could be the result of some ( so far unknown ) astrophysical properties of the sources .
we believe the hypothesis that the neutrino feature be of astrophysical origin should be discarded : the relevant effects are of the order of a couple of days , and neutrinos observed two days before or after a grb could not possibly be grb neutrinos ( unless in - vacuo dispersion takes place ) . if the neutrino feature is confirmed when more abundant data become available we would know that something not of astrophysical origin has been discovered . in this respect
the photon feature is very different .
the size of the effects is between a few and @xmath70 seconds , which may well be the time scale of some mechanisms intrinsic of grbs .
the main reason to be skeptical about the astrophysical interpretation comes from the fact that the content of figure 2 reflects the properties of the @xmath6 : the data points ( those on the main line " ) line up only because we have factored the @xmath6 in the analysis , and the @xmath6 is a form of dependence on redshift which reflects propagation .
so the astrophysical interpretation of the photon feature still requires assuming that at least part of the content of figure 2 is accidental : we can not exclude some mechanism at the grb producing some level of correlation between energy of the photon and difference in time of emission with respect to the gbm peak , but such a mechanism could produce the feature of figure 2 only if accidentally ( on those few data points ) it ended up taking values lending themselves to the sort of @xmath6-dependent analysis which we performed .
so the astrophysical interpretation of the photon feature is possible but must face some issues .
however , we should stress that also the interpretation of the photon feature in terms of the model ( [ mainnew ] ) has to face a challenge connected with grb090510 .
one of 3 photons off the
main line " is a 30 gev photon from the short grb090510 . as discussed above , when taking as working assumption that in - vacuo dispersion actually takes place , these photons off the
main line " should be interpreted as photons emitted not in ( however rough ) coincidence with the first peak of fermi s gbm .
such an interpretation is certainly plausible in general , but the case of the 30 gev photon from grb090510 is a challenge . that 30gev photon was observed @xcite within the half - second time window where most grb090510 photons with energy between 1 and 10 gev were also observed . in light of this
, it is certainly very natural to assume that the 30gev photon could not have accrued an in - vacuo - dispersion effect of more than half a second , travelling from redshift of 0.9 ( the redshift of grb090510 ) , which implies @xmath71 . for @xmath72 , as suggested by the points on the main line " , the in - vacuo - dispersion effect for that 30gev photon should be of more than 15 seconds .
it should have arrived together with that half - second - wide peak of 1 - 10gev photons because of an accidental and strong cancellation between an effect of @xmath73 seconds due to emission - time differences at the source and a 15-second in - vacuo - dispersion effect accrued propagating .
this is certainly possible , but a bit too lucky " for our taste .
we feel the the 30 gev photon from grb090510 poses a very severe challenge for the interpretation of the photon feature in terms of our model ( [ mainnew ] ) , even though all other photons in our data fit so nicely ( [ mainnew ] ) . in connection with this one
should notice that the 30gev photon is the only photon in our sample coming from a short grb ( grb090510 ) .
all other photons in our sample come from long grbs
. if the effect is present for long grbs and absent for short grbs , then the interpretation should be astrophysical .
one can also notice however that grb090510 , with its redshift of 0.9 , is one of the closest grbs relevant for our photon analysis .
all other grbs in our photon analysis , with the exception of grb130427a , are at redshift greater than 1 .
a scenario in which the effect is pronounced only at large redshifts could be of quantum - spacetime origin , but of course would require a quantum - spacetime picture in which the dependence on redshift of the effects is not exactly governed by the function @xmath6 .
more data will soon be available both for our photons and for our neutrinos , so we shall not dwell much on the significance of our findings .
we just stress that surely the false alarm probabilities here derived are small enough to motivate further interest in this type of analyses . particularly for neutrinos a much improved analysis
should become soon possible , since so far icecube only made publicly available their data up to may of 2014 , so at the time of writing this article we know that some additional 2.5 years of data have been collected by icecube but have not yet been publicly released . for photons
our main reference is the fermi telescope , which has been operating since 2008 . in about 8 year of operation
fermi provided 7 grbs contributing to the photon side of our analysis ( see table 2 ) , so we can expect to have roughly one grb per year adding points to our figure 2 .
as stressed above , if the neutrino feature was confirmed it would be very hard to even imagine an astrophysical origin for that feature . for photons instead our intuition , while being open to ultimately finding conclusive evidence of in - vacuo dispersion , presently favors the possibility of a scenario in which the feature is confirmed by additional data but in the end the correct description be given in terms of some properties of the astrophysical sources .
we would welcome feedback from the astrophysics community on the type of mechanisms at the source " that could produce such a feature for photons . on the other hand overall , combining both the neutrino side and the photon side of our analysis , it turn out that the feature is stronger at higher energies and higher redshifts , so we feel that our findings could motivate the development by the quantum - gravity community of models similar to the one of eq.([main ] ) but such that the effects are indeed less pronounced than predicted by eq.([main ] ) at lower energies and/or lower redshifts .
we are very grateful to bo - qiang ma and simonetta puccetti for valuable discussions on some of data here used .
we also gratefully acknowledge conversations with fabrizio fiore and lee smolin .
the work of gr was supported by funds provided by the national science center under the agreement dec- 2011/02/a / st2/00294 .
nl acknowledges support by the european union seventh framework programme ( fp7 2007 - 2013 ) under grant agreement 291823 marie curie fp7-people-2011-cofund ( the new international fellowship mobility programme for experienced researchers in croatia - newfelpro ) , and also partial support from the h2020 twinning project n@xmath74692194 , `` rbi - twinning '' .
s. zhang and b. q. ma , astropart .
* 61 * ( 2014 ) 108 .
h. xu and b. q. ma , astropart .
* 82 * ( 2016 ) 72 .
h. xu and b. q. ma , phys .
b * 760 * ( 2016 ) 602 .
g. rosati , g. amelino - camelia , a. marciano and m. matassa , phys .
d * 92 * , 124042 ( 2015 ) .
x. calmet , s.d.h .
hsu and d. reeb , arxiv:0805.0145 , phys .
* 101 * , 171802 ( 2008 ) .
robinson and f. wilczek , arxiv : hep - th/0509050 , phys .
lett . * 96 * , 231601 ( 2006 ) .
x. calmet , s.d.h .
hsu and d. reeb , phys . rev . * d81 * , 035007 ( 2010 ) .
icecube collaboration : m. g. aartsen _
et al _ , jinst * 9 * , p03009 ( 2014 ) ; m. kadler _ et al _ , nature phys .
* 12 * ( 2016 ) no.8 , 807 ; joel bressieux , _ testing and comparison of muon energy estimators for the icecube neutrino observatory _ , master thesis ( http://lphe.epfl.ch/publications/diplomas/jb.master.pdf ) . | over the last 15 years there has been considerable interest in the possibility of quantum - gravity - induced in - vacuo dispersion , the possibility that spacetime itself might behave essentially like a dispersive medium for particle propagation .
two very recent studies have exposed what might be in - vacuo dispersion features for grb ( gamma - ray - burst ) neutrinos of energy in the range of 100 tev and for grb photons with energy in the range of 10 gev .
we here show that these two features are roughly compatible with a description such that the same effects apply over 4 orders of magnitude in energy .
we also characterize quantitatively how rare it would be for such features to arise accidentally , as a result of ( still unknown ) aspects of the mechanisms producing photons at grbs or as a result of background neutrinos accidentally fitting the profile of a grb neutrino affected by in - vacuo dispersion . |
the total amount of emitted fluorescence light in a shower is a very good approximation to the total number of charged particles @xmath0(@xmath1 ) , where @xmath1 is the atmospheric depth . in this sense
the number of particles at shower maximum can serve as an estimator of the shower energy .
the total energy that goes into electromagnetic charged particles is obtained by integration of the shower longitudinal profile @xmath2 where @xmath3 is the average ionization loss rate and the integral on the right - hand side represents the total track length of all charged particles in the shower projected onto the shower axis . as an alternative proposal @xcite the electromagnetic energy
can also be calculated by using the fluorescence light intensity and the fluorescence efficiency , without the obligation to reconstruct the number of particles as a function of the atmospheric depth .
such approach is taken as a very precise measurement of the primary shower energy because it is supposed to be weakly dependent of the simulation models and the primary particle type .
however , when details of the shower development are taken into account the calorimetric measurement can lead to high systematic uncertainties .
a not less important concern is that the fluorescence efficiency as a function of air pressure , density and humidity is only known up to a certain extent .
according the approach given by equation [ eq : eem ] , the average ionization loss rate is used in the air shower reconstruction and hence the energy spectrum of the electron in the shower must be known via monte carlo simulation .
although the electrons and positrons constitute the majority of the charged particles in a shower and contribute most to the fluorescence light , an important fraction of the shower energy is carried by particles which are invisible to fluorescence telescopes .
such `` missing energy '' is estimated using monte carlo air shower simulation and contributes to the uncertainties involved in this method , being sensitive to the primary composition .
theoretical works have shown the existence of a clear relation between the primary energy and the maximum number of particles in the shower . recently , alvarez - muiz et al .
@xcite have studied the @xmath4 shower quantity as an estimator of the primary shower energy , confirming the efficiency of this technique .
however , telescopes particularities and reconstruction procedures must be considered due to the introduction of biases and fluctuations in the calculation of @xmath4 .
we explore the possibility of estimating the primary shower energy using the @xmath5 shower quantity for the hires @xcite and euso @xcite fluorescence experiments , i.e. , ground and space based experiments , respectively .
the telescopes particularities and the reconstruction procedures are included in our analysis , predicting more realistic results .
the importance of the fluorescence technique in measuring air showers has been demonstrated along the years by the fly s eye and hires collaboration and now is also being successfully used by the pierre auger observatory @xcite , all ground based experiments . at the same time , projects are under development and intend to use the fluorescence approach from space observatories . in such projects , fluorescence telescopes would be installed at the international space station or in satellites which would increase by at least a factor of 10 the aperture reached by the current ground based telescopes . in this work
we study the possibility of using the @xmath5 parameter as an estimator of the primary energy , employing hires and euso telescopes as case studies to test its quality in two different setups : ground and space based experiments . the hires telescope specifications were considered and simulated in complete accordance with @xcite .
the program explained in reference @xcite has been used again to obtain the comparisons between our simulations and the hires data and simulations .
such program scheme was adapted to mimic in details the euso telescope according to the technical configuration given in @xcite .
the longitudinal air shower profiles were generated by corsika @xcite and conex @xcite simulators . from the obtained longitudinal particle profiles
the number of fluorescence photons can be calculated and propagated to the telescopes , being in agreement to the general procedure specified in @xcite .
once the simulation of the shower and the telescopes has been done , the shower longitudinal profile is reconstructed in the standard procedure @xcite .
the number of fluorescence photons measured as a function of time is converted to the number of particles as a function of depth and then a gaisser - hillas profile is fitted .
one of the parameters fitted in the gaisser - hillas function was @xmath4 .
only showers which survived the hires - ii cuts as published in @xcite were used in the following analysis .
the euso collaboration has not defined quality cuts yet and therefore we have imposed very loose ones requiring total path length greater than @xmath6 and greater than 200 g/@xmath7 and @xmath8 in the field of view of the telescope .
we have simulated 100 showers with corsika for energies varying from @xmath9 to @xmath10 ev for the hires analysis and with conex for energies varying from @xmath11 to @xmath12 ev for the euso .
each shower was used 50 times by drawing a different geometry . for the hires calculations
we have randomly distributed the showers with zenith angle smaller than 60@xmath13 over an area with radius of 50 km . for the euso calculations we have drawn core positions in a circle with radius of 430 km and zenith angle smaller than @xmath14 .
[ fig : nmax ] illustrates on the left hand the @xmath4 distribution as simulated by the monte carlo program and the distribution of the same reconstructed quantity on the right hand .
one can easily verify how the detection and reconstruction procedures distort the distribution producing a wider distribution and a shift in the values . any reconstruction method based on @xmath4
should take these biases into account .
[ fig : en : nmax : hi ] shows the relation between the primary energy simulated by the monte carlo schemes for the hires and euso telescopes specifications with the average reconstructed @xmath4 .
the small error bars show the one sigma confidence level for the @xmath4 distribution .
the relation between @xmath4 and energy in fig .
[ fig : en : nmax : hi ] can be well fitted by a straight line @xmath15 the values of _ a _ and _ b _ parameters were determined to be @xmath16 and @xmath17 , respectively , for the hires experiment and @xmath18 and @xmath19 for the euso telescope .
using equations [ eq : en : nmax ] and [ eq : en : nmax ] ( with the a and b parameters as given above ) we simulated a second set of showers and reconstructed the energy using both procedures : a ) the standard integral of the gaisser - hillas profile ( integral ) and b ) the @xmath4 relation ( equation [ eq : en : nmax ] ) .
[ fig : en ] illustrates the error distributions related to the energy reconstruction obtained by using the @xmath4 relation and the integral procedure for the hires experiment , for proton - induced showers at @xmath20 ev . for the hires telescopes
the reconstruction error was calculated to be around 20% at @xmath20 ev reducing to 15% above @xmath21 ev . to this range of energy
the @xmath4 reconstruction showed an average reconstruction error 1.5% smaller than the integral procedure .
[ fig : erro : en ] shows the dependence of the energy reconstruction as a function of energy for the euso telescope . for energies below @xmath22
the reconstruction error related to the @xmath4 method was smaller .
however , for higher energies , the euso telescope is able to detect the entire development of the shower leading to a good fit of the gaisser - hillas at all depths and a better efficiency of the energy reconstruction .
this paper was partially supported by the brazilian agencies cnpq and fapesp .
f. sanchez thanks iag / usp for its hospitality and infn / unimi for the funding support .
most of simulations presented here were carried on a cluster linux tdi , supported by laboratrio de computao cientfica avanada at universidade de so paulo . | the fluorescence technique has been successfully used to detect ultrahigh energy cosmic rays by indirect measurements .
the underlying idea is that the number of charged particles in the atmospheric shower , i.e , its longitudinal profile , can be extracted from the amount of emitted nitrogen fluorescence light .
however the influence of shower fluctuations and the very possible presence of different nuclear species in the primary cosmic ray spectrum make the estimate of the shower energy from the fluorescence data analysis a difficult task .
we investigate the potential of shower size at maximum depth as estimator of shower energy .
the detection of the fluorescence light is simulated in detail and the reconstruction biases are carefully analyzed .
we extend our calculations to both hires and euso experiments .
this kind of approach is of particular interest for showers that are not fully contained inside the field of view of the detector . |
twistor methods in gauge theory have a long history ( summarized in ref . [ p90 ] ) .
a common feature of these methods is that spacetime is replaced by a twistor ( or ambitwistor ) analytic manifold @xmath8 .
equations of motion `` emerge '' ( in terminology of penrose ) from complex geometry of @xmath8 .
the twistor approach turns to be a very useful technical innovation .
for example difficult questions of classical gauge theory , e.g. the ones that appear in the theory of instantons , admit a translation into a considerably more simple questions of analytic geometry of space @xmath8 . on this way classifications theorems in the theory of instantons has been obtained ( see ref .
[ ahdm ] ) .
the quantum theory did not have a simple reformulation in the language of geometry of the space @xmath8 so far .
one of the reasons is that the quantum theory formulated formally in terms of a path integral requires a lagrangian .
classical theory , as it was mentioned earlier , provides only equations of motion whose definition needs no metric . in contrast
a typical lagrangian requires a metric in order to be defined .
thus a task of finding the lagrangian in ( ambi)twistor setup is not straightforward . in this paper
we present a lagrangian for n=3 d=4 yang - mills ( ym ) theory formulated in terms of ambitwistors .
recall that n=3 ym theory coincides in components with n=4 ym theory .
the easiest way to obtain n=4 theory is from n=1 d=10 ym theory by dimensional reduction .
the lagrangian of this ten dimensional theory is equal to @xmath9 in the last formula @xmath10 is a curvature of connection @xmath11 in a principal u@xmath7-bundle over @xmath12 .
an odd field @xmath13 is a section of @xmath14 , where @xmath15 is a complex sixteen dimensional spinor bundle , @xmath16 is the adjoint bundle , @xmath17 is the dirac operator , @xmath18 is a killing pairing on @xmath19 .
the measure @xmath20 is associated with a flat riemannian metric on @xmath12 , @xmath10 are coefficients of the curvature in global orthonormal coordinates .
the n=4 theory is obtained from this by considering fields invariant with respect to translations in six independent directions .
the theory is conformally invariant and can be defined on any conformally flat manifold , e.g .
@xmath21 with a round metric .
e. witten in 1978 in ref .
[ w78 ] discovered that it is possible to encode solutions n=3 supersymmetric ym - equation by holomorphic structures on a vector bundle defined over an open subset @xmath22 in a superquadric @xmath23 .
we shall call the latter a complex ambitwistor superspace . in this description
the action of n=3 superconformal symmetry on the space of solutions is manifest .
symbol @xmath24 denotes dimension of a supermanifold .
more precisely the quadric @xmath1 is defined by equation @xmath25 in bihomogeneous coordinates @xmath26 in @xmath27(@xmath28-even , @xmath29-odd , a symbol @xmath30 in the superscript stands for the dual space ) .
the quadric a is complex supermanifold .
it makes sense therefore to talk about differential @xmath31-forms @xmath32 .
let @xmath33 be a holomorphic vector bundle on @xmath22 .
denote by @xmath34 a differential graded algebra of smooth sections of @xmath35 with coefficients in @xmath36-forms .
let @xmath4 and @xmath37 be two operators corresponding to two holomorphic structures in @xmath33 .
it is easy to see that @xmath38 , where @xmath39 .
the integrability condition @xmath40 in terms of @xmath4 and @xmath41 becomes a maurer - cartan ( mc ) equation : @xmath42 the first guess would be that the space of fields of ambitwistor version of n=3 ym would be @xmath43 , where @xmath33 is a vector bundle on @xmath22 of some topological type .
witten suggested in ref .
[ w03 ] that the lagrangian in question should be similar to a lagrangian of holomorphic chern - simons theory . @xmath44 where @xmath45 is some integral form .
the action ( [ e : gdfsguqd ] ) reproduces equations of motion ( [ e : fdsaf ] ) .
the hope is that perturbative analysis of this quantum theory will give some insights on the structure of n=3 ym .
the main result of the present note is that we give a precise meaning to this conjecture .
introduce a real supermanifold @xmath46 of real superdimension @xmath47 .
it is defined by equation @xmath48 in section [ s : llfddfgg ] we discuss the meaning of reality in superalgebra and geometry .
let @xmath49 be a @xmath50 supermanifold , equipped with a subbundle @xmath51 of the tangent bundle @xmath52 .
we say that @xmath49 is equipped with cr - structure if @xmath51 carries a complex structure defined by fiberwise transformation @xmath53 .
the operator @xmath53 defines a decomposition of the complexification @xmath54 into a direct sum of eigensubbundles @xmath55 .
we say that the cr - structure @xmath53 is integrable if the sections of @xmath2 form a lie subalgebra of @xmath56 under the bracket of vector fields .
a tautological embedding of @xmath0 into the complex manifold @xmath23 induces a cr structure specified by distribution @xmath2 .
properties of this cr structure are discussed in section [ s : qjscmmz ] .
a global holomorphic supervolume form @xmath57 on @xmath23 is constructed in proposition [ p : jsdhd ] . when restricted on @xmath0
it defines a section of @xmath58-a cr integral form .
functorial properties of this form are discussed in section [ s : gufjasklly ] . for any cr - holomorphic vector bundle @xmath33
we define a differential graded algebra @xmath59 .
it is the tangential cr complex .
we equip it with a trace @xmath60 we define a cs - action of the form ( [ e : gdfsguqd ] ) , where we replace an element of @xmath61 by an element of @xmath62 .
the integral is taken with respect to the measure @xmath57 .
we make some assumptions about topology of @xmath33 as it is done in classical twistor theory .
the space @xmath63 is a superextension of @xmath21 ( see section [ s : hfdgjpo ] for details ) .
there is a projection @xmath64 we require that @xmath33 is topologically trivial along the fibers of @xmath65 .
it is an easy exercise in algebraic topology to see that topologically all such bundles are pullbacks from @xmath21 . on @xmath21
unitary vector bundles are classified by their second chern classes .
[ c : gdfsgjj ] suppose @xmath33 is a cr - holomorphic vector bundle on @xmath0 of rank @xmath66 . under the above assumptions
a cs theory defined by the algebra @xmath59 is equivalent to n=3 ym theory on @xmath21 in a principal u@xmath7 bundle with the second chern class equal to @xmath67 . for perturbative computations in ym theory
it is convenient to work in bv formalism .
[ sch00 ] for mathematical introduction and ref .
[ msch06 ] for applications to ym .
[ c : ttc2 ] in assumptions of conjecture [ c : gdfsgjj ] we believe that n=3 ym theory in bv formulation is equivalent to a cs theory defined by the algebra @xmath59 , where the field @xmath68 has a mixed degree
. the following abstract definition will be useful [ d : rtje ] suppose we are given a differential graded algebra @xmath69 with a @xmath70-closed trace functional @xmath71 .
we can consider @xmath72 as a space of fields in some field theory with lagrangian defined by the formula @xmath73 we call it a chern - simons @xmath74 theory associated with a triple @xmath75 we say that two theories @xmath75 and @xmath76 are classically equivalent if there is a quasiisomorphism of algebras with trace @xmath77 see appendix of [ msch05 ] for extension of this definition on a@xmath78 algebras with a trace
. thus the matrix - valued dolbeault complex @xmath79 with a trace defined by the formula @xmath80 would give an example of such algebra .
we shall indicate existence of classical equivalence of n=3 ym defined over @xmath81 and a cs theory defined over @xmath82 , where @xmath22 is an open submanifold of @xmath63 with @xmath83 .
here is the idea of the proof .
we produce a supermanifold @xmath84 and an integral form @xmath45 on it .
we show that a cs theory constructed using differential graded algebra with a trace @xmath85 , associated with manifold @xmath84 is classically equivalent to n=3 ym theory .
we interpret algebra @xmath85 as a tangential cr - complex on @xmath84 .
we shall construct a manifold @xmath84 and algebra @xmath72 in two steps .
here is a description of the steps in more details : * step 1 .
* we define a compact analytic supermanifold @xmath86 and construct an integral form @xmath87 on it in a spirit of [ msch05 ] .
let @xmath88 be the dolbeault complex of @xmath86 .
integration of an element @xmath89 against @xmath87 over @xmath86 defines a @xmath4-closed trace functional on @xmath88 .
we show that the chern - simons theory associated with dga @xmath90 is classically equivalent to n=3 yang - mills theory with gauge group u@xmath7 reduced to a point .
* step 2 . * from the algebra @xmath88 we reconstruct a differential algebra @xmath72 .
the algebra @xmath91 conjecturally encodes full n=3 yang - mills theory with gauge group u@xmath7 in a sense of definition [ d : rtje ] . if we put aside the differential @xmath70
, @xmath72 is equal to @xmath92 .
the integral form we are looking for is equal to @xmath93 , where @xmath94 is equipped with global coordinates @xmath95 .
the manifold @xmath84 is a cr submanifold . we identify it with an open subset of @xmath0 .
finally we would like to formulate an unresolved question .
restriction of a holomorphic vector bundle @xmath33 over @xmath22 on @xmath96 defines a cr - vector bundle over the intersection .
is it true that any cr - holomorphic vector bundles can be obtained this way ?
the answer would be affirmative if we impose some analyticity conditions on the cr structure on @xmath33 .
presumably super levi form will play a role in a solution of this problem .
it is tempting to speculate that there is a string theory on @xmath23 and @xmath0 defines a d - brane in it .
we need to say few words about the structure of this note . in section [ s : dfdrf ]
we make definitions and provide some constructions used in formulation of conjectures ( [ c : gdfsgjj ] ) and ( [ c : ttc2 ] ) . in section [
s : main ] we give a geometric twistor - like description of n=3 ym theory reduced to a point ( step 1 ) . in section [ s : ffkewx ]
we do the step 2 .
appendix contains some useful definitions concerning reality in superalgebra and cr - structures .
in this section we shall show that the space @xmath0 is homogeneous with respect to the action of a real form of n=3 superconformal algebra @xmath97 . here
we also collected facts that are needed for coordinate - free description of space @xmath0 in terms of lie algebras of symmetry group and isotropy subgroup . in this section
we describe a graded real structure on @xmath97 .
it will be used later in construction of cr - structure on real super - ambitwistor space .
the reader might wish to consult section [ s : llfddfgg ] for the definition of a graded real structure . there
the reader will find explanation of some of our notations . by definition @xmath97
is a super lie algebra of endomorphisms of @xmath98 .
symbol @xmath99 stands for the parity change .
this lie algebra consists of matrices of a block form @xmath100 with @xmath101 .
elements @xmath102 belong the even part @xmath103 , elements @xmath104 to the odd @xmath105 . in the following a symbol @xmath106 will stand for a lie algebra defined over a field @xmath107 .
if the field is not present it means that the algebra is defined over @xmath108 .
the same applies to lie groups .
let @xmath109 be a complex super lie algebra . by definition a map @xmath110 that defines a graded real structure on on super lie algebra @xmath109 if @xmath110 is a homomorphism : @xmath111=[\rho(a),\rho(b)]$ ] . in ref
[ man ] yu .
i. manin suggested several definitions of a real structure on a ( lie ) superalgebra . in notations of [ man ]
these definitions are parametrized by a triple @xmath112 .
our real structure correspond to the choice @xmath113 .
the reader will find a complete classification of graded real structures on simple lie algebras in the work [ serg ] .
define a matrix @xmath53 as : @xmath114 where @xmath115 is a @xmath116 identity matrix .
a map @xmath110 is defined as @xmath117 the identity @xmath118 is defined in the appendix in definition [ d : gdfjgdf ] . ] is a corollary of equation @xmath119 .
it is useful to analyze the lie subalgebra @xmath120 of real points in @xmath121 . due to
( [ e : gfjeiuq ] ) we have @xmath122 .
to identify @xmath123 we interpret @xmath124(whose algebra of endomorphisms is @xmath125 ) as a two - dimensional quaternionic space @xmath126 .
let @xmath127 be the standard @xmath128-basis in quaternions , @xmath129 be an @xmath130-basis in @xmath126 .
the space @xmath131 has a complex structure defined by the right multiplication on @xmath132 .
the right multiplication on @xmath133 defines an @xmath132-antilinear map . in a @xmath108-basis
@xmath134 a matrix of right multiplication on @xmath133 is equal to @xmath53 . from this
it is straightforward to deduce that @xmath135 .
let @xmath49 be a @xmath50 supermanifold with a tangent bundle @xmath52 .
let @xmath136 be a subbundle equipped with a complex structure @xmath53 .
this data defines a ( nonintegrable ) cr - structure on @xmath49 .
there is a decomposition in superscript denotes complexification . ]
a cr - structure @xmath138 is integrable if a space of sections of @xmath2 is closed under commutator . in this case
we also say that @xmath2 is integrable .
let @xmath139 denote the underlying manifold of supermanifold @xmath49 .
if @xmath49 is a real submanifold of a complex supermanifold @xmath140 then at any @xmath141 the tangent space @xmath142 to @xmath49 contains a maximal complex subspace @xmath143 .
if @xmath144 is constant along @xmath49 then a family of spaces @xmath51 defines an integrable cr - structure . in our case the manifold @xmath46 is defined by equation ( [ e : dfsad ] ) .
denote by @xmath145 an affine supergroup with lie algebra @xmath146 equal to @xmath97 see ref .
[ pel ] . ] .
we will show later that @xmath0 is a homogeneous space of a real form of @xmath145 described above .
the induced cr - structure is real - analytic and homogeneous with respect to the group action .
a cr - structure on a supermanifold enables us to define an analog of dolbeault complex .
suppose a supermanifold @xmath49 carries a cr structure @xmath147 .
a space of complex 1-forms @xmath148 contains a subspace @xmath149 of forms pointwise orthogonal to @xmath2 .
it is easy to see that @xmath2 is integrable iff the ideal @xmath150 is closed under @xmath70 . define a tangential cr - complex @xmath151 to be @xmath152 .
a vector bundle @xmath33 is cr - holomorphic if the gluing cocycle @xmath153 satisfies @xmath154 . in such case
we can define a @xmath33-twisted cr - complex @xmath155 .
denote by @xmath156 an operation of complex conjugation .
define an antilinear map @xmath157 a map @xmath158 coincides with the real structure @xmath110 .
let us think about lhs of equation ( [ e : kdsj ] ) as a quadratic function associated with an even bilinear form @xmath159 .
it is easy to see that lhs of equation ( [ e : dfsad ] ) is equal to @xmath160 . naively thinking the centralizer of operator @xmath161 would precisely be the real form of @xmath162 and it would preserve equations ( [ e : kdsj ] ) and ( [ e : dfsad ] ) .
the problem is that we can not work pointwise in supergeometry .
instead we consider equations ( [ e : kdsj ] , [ e : dfsad ] ) as a system of real algebraic equations .
we interpret them as a system of sections of some line bundles on @xmath163 manifold @xmath164 ( see section [ s : llfddfgg ] for discussion of reality in supergeometry ) .
the space @xmath49 carries a canonical graded real structure @xmath110 , that leaves the space of equations invariant .
the @xmath110-twisted diagonal action of @xmath97 also leaves the equations invariant .
the graded real structure induces a graded real structure @xmath110 on @xmath165 and makes a supermanifold @xmath0 an algebraic graded real supermanifold .
we define a space @xmath166 as a homogeneous space of a real supergroup @xmath167 . in this section
we establish an isomorphism @xmath168 . in fig .
( [ f : gdfgdjh ] ) the reader can see a graphical presentation of some matrix @xmath169 .
the isotropy subalgebra @xmath170 of a base point in the space @xmath166 is defined as a linear space of matrices whose nonzero entries are in the darkest shaded area of a matrix in fig .
( [ f : gdfgdjh ] ) .
[ ht ] subspace @xmath170 is a @xmath110-invariant subalgebra .
direct inspection .
let @xmath72 be an algebraic subgroup of @xmath145 with a lie algebra @xmath171 .
the space @xmath166 carries a homogeneous cr structure ( see section [ s : gfgjheq ] for related discussion ) .
define a subspace @xmath172 as a set of matrices with nonzero entries in gray and dark gray areas in fig .
( [ f : gdfgdjh ] ) .
subspace @xmath172 is a subalgebra .
it satisfies @xmath173 direct inspection .
let @xmath174 denote an algebraic subgroup with lie algebra @xmath175 .
a complex supermanifold @xmath176 has an explicit description .
equation ( [ e : kdsj ] ) is preserved by the action of @xmath145 .
there is a @xmath145-equivariant isomorphism @xmath177 .
we can identify the quadric @xmath23 with the space of partial flags @xmath178 in as it is done in purely even case ( see [ gh ] for example ) .
a spaces @xmath23 is a connected component of the flag space containing the flag @xmath179 with @xmath180 and @xmath181 . this flag can be interpreted as a pair of points @xmath182 , @xmath183 .
the condition ( [ e : yhggb ] ) is equivalent to ( [ e : kdsj ] ) .
let us choose a standard basis @xmath184 of @xmath178 such that the parities of elements are @xmath185 , @xmath186 . in this notations
the standard flag @xmath187 has the following description : @xmath188 the flag defines a point in the space @xmath23 .
it is easy to compute a shape of the matrix of an element from the stabilizer @xmath189 of @xmath187 .
the following picture is useful @xmath190{$\mathfrak { gl}(2|3)$}}\put(1276,-1336){\makebox(0,0)[lb]{$\mathbb{c}^{*2|3}$}}\put(2476,-1336){\makebox(0,0)[lb]{$\mathbb { c}$}}\put(451,539){\makebox(0,0)[lb]{$\mathbb{c}$}}\put(2401,539){\makebox(0,0)[lb]{$\mathfrak{rad}$}}\put(451,-1336){\makebox(0,0)[lb]{$\mathbb{c}^{1|0}$}}\put(451,-436){\makebox(0,0)[lb]{$\mathbb{c}^{2|3}$}}\end{picture } } \ ] ] the lie algebra @xmath191 of stabilizer @xmath192 is formed by matrices with zero entries below thick solid line in picture ( [ e : fgojkxcz ] ) .
conjugating with a suitable permutation of coordinates @xmath193 we see that @xmath194 .
a transformation @xmath161 defined in equation ( [ e : gjhsirpqqa ] ) acts on the space of flags . by definition
an @xmath161-invariant flag belongs to the subvariety @xmath0 .
a direct inspection shows that the nonzero entries of the matrix of an element of stabilizer are located in the darkest shaded area on fig .
( [ f : gdfgdjh ] ) . the manifold @xmath195 fibers over @xmath196 with connected fibers
. thus @xmath0 is connected . from this and a simple dimension count
we conclude that subvariety @xmath0 coincides with @xmath166 .
identification of cr structure also follows from this .
the lie algebra @xmath97 contains a subalgebra @xmath197 .
the elements of this subalgebra have nonzero entries in the darkest area on fig ( [ f : gdfgdjh ] ) and also spots marked by @xmath30 .
this algebra is invariant with respect to the real structure @xmath110 .
denote by @xmath198 an algebraic subgroup of @xmath145 with lie algebra @xmath197 .
the quotient @xmath199 is a supermanifold with @xmath200 . indeed
the real points @xmath201 of the group @xmath202 are conjugated to quaternionic matrices of the form @xmath203 .
thus the quotient space @xmath204 is isomorphic to @xmath21 .
denote by @xmath65 projection @xmath205 an easy local exercise with lie algebras reveals that the fibers of projection @xmath65 are cr - holomorphic and are isomorphic to @xmath206 . the following direct geometric description of ambitwistor space will be useful .
let @xmath49 be a @xmath50 4-dimensional riemannian manifold .
a metric @xmath207 defines a relative quadric ( the ambitwistor space ) in the projectivisation of a complexified tangent bundle @xmath208 . by construction
there is a projection @xmath209 .
the space @xmath210 carries a cr structure ( it could be nonintegrable ) . indeed a fiber of the distribution @xmath2 at a point @xmath211 is a direct sum of the holomorphic tangent space to the fiber through @xmath212 and a complex line in @xmath213 spanned by @xmath212 . from the point of view of topology
the space @xmath210 coincides with a relative grassmannian of oriented 2-planes in @xmath214 .
a constructed complex distribution depends only on conformal class of the metric . from this
we conclude that @xmath215 ( @xmath21 has a round metric induced by the standard embedding into @xmath216 ) is a homogeneous space of @xmath217 , @xmath218 and the cr structure is integrable
. an appropriate super generalization of this construction is as follows .
we have an isomorphism @xmath219 in the last formula @xmath220 are complex two - dimensional spinor bundles on @xmath49 ( we assume that @xmath49 has a spinor structure ) . the isomorphism @xmath221 is defined by clifford multiplication .
let @xmath52 be a 3-dimensional linear space .
this vector space will enable us to implement n@xmath222 supersymmetry . to simplify notations we keep @xmath223 for the pullback @xmath224 .
define a split , holomorphic in odd directions on this . ]
supermanifold @xmath225 associated with a vector bundle @xmath226 over @xmath210 . to complete the construction
we define a superextension of the cr structure .
introduce odd local coordinates @xmath227 ( @xmath228 ) on fibers of @xmath226 .
we decompose local complex vector fields @xmath229 in a local real basis @xmath230 as @xmath231 .
the odd part of the cr distribution @xmath2 is locally spanned by vector fields @xmath232 this construction of a superextension of ordinary cr structure depends only on the conformal class of the metric .
it is convenient to formally add the complex conjugate odd coordinates .
this way we get @xmath233 , equipped with a graded real structure . as in the even case
the symmetry analysis allows to identify cr - space @xmath234 with @xmath0 .
a tangent space @xmath235 to @xmath23 at a point fixed by @xmath175 is formed by elements with nonzero entries below thick solid line on the picture ( [ e : fgojkxcz ] ) .
it decomposes into a sum @xmath236 of irreducible @xmath237 representations .
the elements @xmath238 stand for matrix coordinate functions on the linear space @xmath235 ( coordinates @xmath41 are even , @xmath239 are odd ) .
[ p : jsdhd ] an element @xmath240 belongs to @xmath241 .
it is invariant with respect to the action of @xmath174 .
simple weight count .
we spread a generator of @xmath241 by the action of @xmath145 over @xmath23 and form a @xmath145-invariant section @xmath57 of @xmath242
as a preliminary step in construction of the superspace @xmath84 we introduce a `` holomorphic '' manifold @xmath243 and an integral form on it ( see section [ e : gfgdfgjr ] for the explanation of quotation marks ) .
the form defines a functional @xmath71 on dolbeault complex of this manifold .
we prove that the cs theory constructed by the triple @xmath244 is classically equivalent to n=3 ym theory with gauge group u@xmath7 reduced to a point .
a manifold @xmath246 is a deformation of a more simple manifold @xmath245 . in this section
we give relevant definitions concerning @xmath245 .
denote a product @xmath247 by @xmath248 .
it has two projections @xmath249 .
let @xmath250 denote the dual to the hopf line bundle over @xmath251 .
the picard group of @xmath248 is @xmath252 .
it is generated by the classes of line bundles @xmath253 , @xmath254 that can serve as coordinates in @xmath255 .
let @xmath256 denote a line bundle @xmath257 .
* convention * we denote by @xmath258 the cohomology of ( super)manifold @xmath259 with coefficients in a vector bundle @xmath33 .
it can be computed as cohomology of dolbeault complex @xmath260 .
it is tacitly assumed that in the section [ s : main ] the omitted argument @xmath259 in @xmath260 implies @xmath261 .
if @xmath33-argument is missing we assume that @xmath262 .
denote by @xmath263 symmetric and exterior algebras of a vector space ( bundle ) .
denote by @xmath264 a line bundle isomorphic to @xmath265 .
we construct a vector bundle @xmath187 over @xmath248 as a direct sum : @xmath266 as before @xmath52 is a three dimensional vector space .
the reader may have noticed that the manifold @xmath248 has also appeared as a fiber of projection ( [ e : sdsfsew ] ) .
we shall see this is not accidental . in this section
we devise an infinitesimal deformation of a complex structure on @xmath245 .
this deformation will be promoted to the actual deformation which we denote by @xmath246 .
the algebra @xmath88 from the introduction is equal to @xmath267 .
we construct on @xmath246 we construct an integral form that will enable us to define a functional @xmath268 . the manifold @xmath245 is a complex split supermanifold .
the dolbeault complex @xmath269 is defined on supermanifold @xmath245 , considered as a graded real supermanifold .
the complex @xmath269 contains as a differential subalgebra the dolbeault complex @xmath270 .
differential algebras @xmath271 and @xmath270 are quasiisomorphic .
the same as a proof of proposition [ e : fdsafjh ] .
the canonical line bundle @xmath272 is equal to @xmath273 .
there is a nontrivial cohomology class- the `` fundamental '' class : @xmath274 .
we interpret @xmath275 as a sheaf of local holomorphic differentiations of @xmath245 in direction of @xmath276 .
a representative @xmath277 ( @xmath278 are local coordinates on @xmath248 ) of the class @xmath279 $ ] can be extended to a differentiation of @xmath271 .
the main properties of @xmath280 are : + @xmath281 it is a differentiation of @xmath271 , + @xmath282 equation @xmath283 holds .
+ all of them are corollaries of @xmath4-cocycle equation for @xmath239 .
the operator @xmath284 defines a new `` holomorphic '' structure on @xmath245 .
is an element of @xmath285 ( @xmath286 is a holomorphic tangent bundle ) , whereas in our case @xmath287 .
we however continue to use a traditional wording and call it a deformation of a complex structure , though a more precise term would be deformation of the dolbeault algebra @xmath288 .
this algebra in our approach becomes a substitute for the underlying manifold .
] this new complex manifold will be denoted by @xmath289 .
the manifold @xmath245 is calabi - yau . by this
we mean that @xmath290 is trivial .
indeed the determinant line bundle of @xmath187 is equal to @xmath291-is trivial .
it implies that the bundle @xmath292 admits a nonvanishing section @xmath293 .
this section is @xmath294-invariant .
the action of @xmath295-the unipotent subgroup of borel subgroup @xmath296 on the large schubert cell of @xmath248 is transitive and free .
hence the section of @xmath297 in @xmath298-coordinates is @xmath299 where @xmath300 are @xmath298-invariant coordinates on the odd fiber .
the section @xmath293 is in the kernel of @xmath284 by construction .
we can construct on manifold @xmath245 a global holomorphic integral @xmath301-form the way explained in remark ( [ e : hhghssa ] ) . in our case
it is equal to @xmath302 .
the form @xmath303 is @xmath284 closed nontrivial integral @xmath304-form on the underlying real graded @xmath245
. direct inspection in local coordinates .
an integral form @xmath87 defines a @xmath305-closed trace on @xmath306 @xmath307 by definition an a@xmath78 algebra algebra is a graded linear space , equipped with a series of maps @xmath308 of degree @xmath309 that satisfy quadratic relation : @xmath310 where @xmath311 , and @xmath312 .
in particular , @xmath313 .
[ e : gfdadfjj ] suppose we have an a@xmath78 algebra @xmath72 equipped with a projector @xmath314 .
a homotopy @xmath51 such that @xmath315 can be used as an input data for construction of a new a@xmath78 structure on @xmath316 ( see [ kad],[markl ] for details ) .
the homotopy @xmath51 is not unique .
the resulting a@xmath78 algebras will have different multiplications , depending on @xmath51 .
all of them will be a@xmath78 equivalent .
an additional structure on @xmath72 helps to fix an ambiguity in a choice of @xmath51 . in our case algebra
@xmath72 is a dolbeault complex of a manifold with an operator @xmath314 being an orthogonal projection on cohomology .
if the manifold is compact , khler and a @xmath317-homogeneous there is natural choice of @xmath51 : @xmath318 .
the operator @xmath319 @xmath320 are build by a @xmath317-invariant metric .
the operator @xmath320 is equal to @xmath321 on @xmath322 and equal to identity on @xmath323 .
[ e : gfdadfjjdsfd ] a construction described in remark ( [ e : gfdadfjj ] ) admits a generalization .
suppose an a@xmath78 algebra @xmath72 has a differential @xmath70 that is a sum of two anticommuting differentials @xmath324 and @xmath325 .
assume that @xmath326 and a composition @xmath327 is a nilpotent operator .
then @xmath316 carries a structure an a@xmath78 algebra quasiisomorphic to @xmath72 .
the same statement is true for a@xmath78 algebras with a trace .
the proof goes along the same lines as in ref .
[ markl ] , but we allow two - valent vertices . technically it is more convenient to work not with algebra @xmath328 but with a quasiisomorphic subalgebra @xmath329 . in application of the constructions from the remarks ( [ e :
gfdadfjj ] , [ e : gfdadfjjdsfd ] ) we choose @xmath314 to be an orthogonal projector from the hodge theory , corresponding to @xmath330-invariant metric on @xmath248 .
we also use a decomposition @xmath331 .
the algebra of cohomology of @xmath332 carries an a@xmath78-algebra structure
. we denote it by @xmath333 .
denote by @xmath334 a quasiisomorphism @xmath335 .
we shall describe some properties of @xmath336 .
let @xmath337 , @xmath338 be spinor representations of @xmath294 .
the vector representation @xmath339 is equal to @xmath340 .
the differential @xmath239 induces a differential @xmath341 $ ] on @xmath336 .
the ghost grading of the group @xmath342 is equal to @xmath343 , the additional grading is equal to @xmath344(preserved by @xmath4 and @xmath239 ) .
we used a nonstandard ghost grading that differs from the one used in physics on shift by one .
in particular the ghost grading of the gauge ( labeled by @xmath339 ) and spinor ( labeled by spinors @xmath337 , @xmath338 ) and matter ( @xmath294 action is trivial ) fields is equal to one . in the table below you will find the field content ( representation theoretic description ) of @xmath336 : @xmath345 the groups @xmath346 and @xmath347 are contracting pairs , they are killed by the differential @xmath341 $ ] and should be considered as auxiliary fields in the related cs theory .
an a@xmath78 algebra @xmath336 besides differential @xmath341 $ ] and multiplication has higher multiplication on three arguments ( corresponding to cubic nonlinearity of ym equation ) .
however operations in more then three arguments are not present . this can be deduced from homogeneity of @xmath4 and @xmath239 with respect to the additional grading .
finally representation theory fixes structure maps up to a finite number of parameters .
the integral @xmath71 defines a nonzero map @xmath348 .
presumably it is possible to complete this line of arguments to a full description of multiplications in @xmath336 .
we prefer do it indirectly through the relation to berkovits construction [ berk ] .
let @xmath12 be a linear space , equipped with a positive - definite dot - product . denote by @xmath15 an irreducible complex spinor representation of orthogonal group @xmath349 .
denote by @xmath350 coefficients of the nontrivial intertwiner @xmath351 in some basis of @xmath15 and an orthonormal basis of @xmath352 .
we assume that @xmath352-basis is real . on a superspace @xmath353
we define a superfunction ( lagrangian ) @xmath354[a_i , a_j])+\sum_{\alpha\beta i}tr(\gamma_{\alpha\beta}^i[a_i,\chi^{\alpha}]\chi^{\beta})\ ] ] @xmath355 is a collection of antihermitian matrices labeled by the basis of @xmath352 .
similarly odd matrices @xmath356 are labeled by the basis of @xmath15 .
this can be considered as a field theory , obtained from d=10 , n=1 ym theory by reduction to zero dimensions .
we call it ikkt after the paper [ ikkt ] where it has been studied .
ikkt theory has a gauge invariance - invariance with respect to conjugation .
a bv version of ikkt coincides with a cs theory associated with an a@xmath78 algebra @xmath357 that we shall introduce presently .
an a@xmath78 algebra @xmath357 can be considered as vector space spanned by symbols @xmath358 with operations @xmath359 ( multiplication ) , @xmath360(massey product ) defined by the following formulas : @xmath361 all other products are equal to zero .
an element @xmath362 is a unit .
all operations @xmath363 with @xmath364 vanish .
the algebra carries a trace functional @xmath365 equal to one on @xmath366 and zero on the rest of the generators .
it induces a dot - product by the formula @xmath367 , compatible with @xmath368 . by definition
an a@xmath78 algebra has a grading ( we call it a ghost grading ) such that operation @xmath369 has degree @xmath309 .
an a@xmath78 algebra might also have an additional grading such that all operations have degree zero with respect to it .
see [ msch05 ] for details on gradings of @xmath370 .
[ p : bbncnsq ] a differential graded algebra with a trace @xmath371 is quasiisomorphic to @xmath357 .
we shall employ methods developed in [ msch05 ] . recall that the manifold of pure spinors in dimension ten is equal to @xmath372 . as a complex manifold it is defined as the space of solutions of homogeneous equations @xmath373 , where @xmath374 are homogeneous coordinates on @xmath375 .
a space @xmath375 is the projectivisation of irreducible complex spinor representation of @xmath376 .
we denote by @xmath377 the restriction on @xmath378 of the twisted tangent bundle @xmath379 .
denote by @xmath72 the coordinate algebra @xmath380/\gamma_{\alpha\beta}^i\lambda^{\alpha}\lambda^{\beta}$ ] of @xmath378 .
denote by @xmath381 the koszul complex @xmath382 $ ] with differential @xmath383 .
the algebra @xmath381 can be equipped with various gradings .
the cohomological grading is defined on generators as @xmath384 , @xmath385 ; the grading by the degree ( or homogeneous grading ) is @xmath386 .
the differential has degree one with respect to @xmath387 and zero with respect to @xmath388 . as a result
the cohomology groups are bigraded : @xmath389 , where @xmath132 corresponds to @xmath387-grading , @xmath133 corresponds to @xmath388 . in [ msch05 ]
we proved that @xmath390 .
in fact we proved that the identification map is a quasiisomorphism of differential graded algebras with a trace . in the language of supermathematics
we may say that cohomology @xmath391 is dolbeault cohomology of a split supermanifold @xmath392 .
in the course of the proof of quasiisomorphism we have identified the algebra of functions on the fiber of projection @xmath393 over a point @xmath394 with cohomology of algebra @xmath395,d)$ ] where @xmath396 .
an analog of the complex @xmath397 can be defined for any subscheme @xmath22 of @xmath378 as a tensor product @xmath398 $ ] .
the algebra @xmath399 is equal to @xmath400 .
the cohomology of @xmath401 is equal to @xmath402 if @xmath403 is trivial on @xmath22 .
we proved in [ msch06 ] that the algebra @xmath381 with berkovits trace @xmath365 is quasiisomorphic to algebra @xmath370 .
we need to present a useful observation from [ msch05 ] .
let us decompose a set @xmath404 into a union @xmath405 such that @xmath406 is a regular sequence .
then the algebras @xmath407 and @xmath408,d)$ ] are quasiisomorphic .
the following construction has been described in [ msch05].the spin representation @xmath15 of @xmath409 splits after restriction on @xmath410 into @xmath411 .
we choose coordinates on @xmath412- equal to @xmath413 .
they form a regular subsequence of @xmath374 .
the manifold corresponding to @xmath414 is equal to @xmath415 .
the intersection is isomorphic to @xmath416 .
the algebra of homogeneous functions @xmath417 on @xmath416 is generated by @xmath418 .
the relations are @xmath419 in the formula @xmath420 stands for a row of @xmath116 minors of @xmath421 matrix .
we plan to follow almost the same method of construction of supermanifold as for the koszul complex of pure spinors .
denote by @xmath207 a projection @xmath422 we fix a point @xmath423 .
the algebra @xmath424 is isomorphic to + @xmath425/(p_iu^i)$ ] .
the algebra @xmath426 is isomorphic to @xmath427,d\ ] ] with a differential @xmath428 the cohomology of this differential is equal to @xmath429=\lambda[\pi_1,\dots,\pi_3,\nu^1,\dots,\nu^3,\theta]\ ] ] the induced a@xmath78 algebra structure on cohomology has no higher multiplications .
the element @xmath430 is represented by a cocycle @xmath431 .
the linear space @xmath432 coincides with the fiber @xmath433 of vector bundle @xmath187 ( [ e : sjs ] ) .
the main distinction between this computation and a computation with pure spinors is that we encountered a noncanonical a@xmath78 morphism @xmath434 $ ] , which could be not a homomorphism of associative algebras .
recall that we viewed the cohomology of @xmath435 as functions of the fiber of projections @xmath436 .
we have a natural identification of fibers over different patches of @xmath378 .
it gives us a consistent construction of a split manifold @xmath392 . in case of the manifold
@xmath248 if we ignore the issues related to ambiguities of choice of morphism @xmath437 tt is not hard to see that an isomorphism of fibers @xmath438 can be extended to a @xmath439 equivariant isomorphism of the vector bundles @xmath438 ( use homogeneity of both vector bundles with respect to @xmath440 action ) . this way we recover the manifold @xmath245 . in reality
when we try to glue rings of functions on different patches the structure isomorphisms will be a@xmath78-morphisms .
we may claim on general grounds that we get an a@xmath78 structure on a space of ech chains of @xmath441 .
this structure can be trivialized by a twist on a local a@xmath78-morphism ( reduced to the standard multiplication in grassmann algebra ) on every double intersection @xmath442 of patches ( if @xmath443 is sufficiently small ) .
an ambiguity in a choice of such twist leads to appearance of ech 2-cocycle @xmath444 with values in infinitesimal ( not a@xmath78 ) transformations of the fiber @xmath245 . in the dolbeault picture
this cocycle corresponds to @xmath239 .
finally we use ech - dolbeault equivalence .
this proves the claim . the algebra @xmath336 carries a differential @xmath445 $ ] . the minimal model of @xmath336 ( by definition
it is a quasiisomorphic a@xmath78 algebra without a differential ) constructed for an obvious homotopy of differential @xmath445 $ ] is quasiisomorphic to @xmath370 .
if we ask for a quasiisomorphism to be compatible with all gradings that exist on both algebras this quasiisomorphism is an isomorphism . from this
is is quite easy to recover all multiplications in the algebra @xmath336 .
a manifold @xmath84 with an integral form is constructed in this section . in this section
we construct an algebra @xmath85 .
it will be a linking chain between ym theory and ambitwistors .
we construct a manifold @xmath84 as a direct product @xmath446 . intuitively speaking the algebra
@xmath267 carries all information about ym theory reduced to a point , whereas algebra @xmath447 contains similar information about the space @xmath94 .
the idea is that a tensor product @xmath448 with a suitably twisted differential will contain all information about 4-d ym theory .
later we will interpret the same complex as tangential cr complex on the manifold @xmath84 .
the linear space @xmath449 has coordinates @xmath450 .
the vector space @xmath339 has an @xmath451 action compatible with decomposition ( [ e : gdfhd ] ) .
it is induced from the @xmath330 action on @xmath94 .
define a differentiation of @xmath452 as follows .
there is an @xmath294 equivariant isomorphism @xmath453 .
the line bundle @xmath265 is generated by its global sections .
we have a short exact sequence : @xmath454 where @xmath339 is considered as a trivial vector bundle with fiber @xmath339 .
the differentiation @xmath455 of @xmath456 is equal @xmath457 .
we interpret @xmath458 as global sections of @xmath459 .
recall that the differential @xmath284 in @xmath328 is equal to @xmath305 .
it becomes clear from explicit computation of cohomology that coefficients @xmath460 in @xmath461 are @xmath4-exact .
choose @xmath462 such that @xmath463 .
define a differentiation @xmath464 of @xmath452 as @xmath465 and zero on the rest of the generators .
it follows from our construction that @xmath466 satisfies @xmath467 .
denote @xmath468 . by definition
the integral form on manifold @xmath469 is : @xmath470 @xmath45 is invariant with respect to @xmath284 , @xmath471 and therefore with respect to @xmath472 .
direct inspection .
[ p : jfjhsg ] denote @xmath473 .
we equip algebra @xmath474 with a trace @xmath475 the cs theory constructed by the triple @xmath476 is classically equivalent to n=3 euclidean ym with a gauge group u@xmath7 .
the equivalence should hold also on a quantum level .
we shall only outline the basic ideas .
a precise mathematical statement is about quasiisomorphism of certain a@xmath78 algebras .
one of them is @xmath85 .
the reader should consult [ msch06 ] for information about a@xmath78 algebra with a trace corresponding to n=3 ym theory .
suppose @xmath477 is a quasiisomorphism of two a@xmath78 algebras .
let @xmath478 be an associative algebra .
then we have a quasiisomorphism of tensor products @xmath479 .
let @xmath41 be a solution of mc equation in @xmath480 ( the reader may consult [ msch05 ] for the definition ) .
the map @xmath334 transports solution @xmath41 to a solution @xmath481 of mc equation in @xmath482 . in a formal interpretation of structure maps of a@xmath78 algebra @xmath72 as the taylor coefficients of a noncommutative vector field on noncommutative space @xmath483 solution @xmath41 of mc equation corresponds to a zero of the vector field .
we can expand the vector field into series at @xmath41 and get some new a@xmath78 algebra .
this construction is particularly transparent in case of a dga .
a solution of mc equation defines a new differential @xmath484 $ ] .
it corresponds to a shift of a vacuum in a physics jargon .
denote by @xmath485 an a@xmath78 algebra constructed by the element @xmath41 .
it is easy to see that the map @xmath334 defines ( under some mild assumptions in @xmath41 ) a quasiisomorphism @xmath486 denote by @xmath487 an algebra of differential operators on @xmath94 .
we would like to apply construction from the previous paragraph to the tensor products @xmath488 and @xmath489 .
we can interpret @xmath490 as a solution of mc equation for algebra @xmath267 with coefficients in @xmath487 .
the quasiisomorphism @xmath334 maps @xmath490 into an element @xmath491 , where @xmath492 is a basis of @xmath493 .
the rest is a matter of formal manipulations .
it is straightforward to see that @xmath494 is an a@xmath78 mathematics counterpart of ym equation where the gauge potential , spinors , matter fields are having their coefficients not in functions on @xmath94 but in @xmath487 .
this is not precisely what we have hoped to obtain .
we shall address this issue presently .
suppose an associative algebra @xmath478 contains a subalgebra @xmath495 .
the previous construction has a refinement . the noncommutative vector field on a space @xmath496 corresponding to a@xmath78 algebra @xmath480 is tangential to a noncommutative subspace @xmath497 ( because @xmath495 is closed under multiplication ) .
we say that a solution of mc @xmath498 is compatible with @xmath495 if the vector field defined by the algebra @xmath499 is tangential to the space @xmath500 .
this is merely another way to say that a linear space @xmath501 is a subalgebra of @xmath502 . if @xmath41 is compatible with @xmath495 the map @xmath334 ( under some mild assumptions on @xmath41 ) induces a quasiisomorphism @xmath503 .
we apply this construction to subalgebra @xmath504 for which mentioned above condition on @xmath490 is met .
suppose in addition that algebras @xmath505 have a traces and a morphism @xmath477 is compatible with the traces .
assume moreover that the induced a@xmath78 structure @xmath506 is compatible with a trace .
then the induced morphism @xmath507 is compatible with traces . in our case
the operator @xmath471 preserves the integral form @xmath45 and the above conditions are met . the cs theory associated with the algebra
@xmath508 has the following even part of the lagrangian @xmath509,k_{\alpha}^{\beta}>+<k_{\alpha}^{\beta},k^{\alpha}_{\beta}>\ ] ] the theory besides of the gauge field corresponding to connection @xmath510 in a principle u@xmath7 bundle , matter fields @xmath511 contains an auxiliary field @xmath512 .
this theory is equivalent to n=3 ym ( the odd parts of the lagrangians coincide ) . in this section
we give a geometric interpretation of the algebra @xmath513 .
fibers of projection @xmath514 have holomorphic structure
. denote @xmath515 a bundle of @xmath65-vertical vector fields .
we define a distribution @xmath33 as @xmath516 choose a linear basis @xmath517 .
define @xmath230 the differentiations in the direction of @xmath518 .
restrict a map @xmath478 from short exact sequence ( [ q : wert ] ) on @xmath519 , then @xmath520 is a set of holomorphic sections of @xmath264 . for any point @xmath423
we have a subspace @xmath521 spanned by @xmath522 a union of such subspaces defines a complex distribution @xmath523 on @xmath524 . define an integrable distribution @xmath525 .
the reader can see that the cr structure on the space @xmath524 literally coincides with the cr structure on ambitwistor space @xmath526 defined in section [ s : hfdgjpo ] . in light of this identification an element @xmath430 ( a local coordinate on @xmath276 )
can be interpreted as a local cr - form with nonzero values on @xmath527 .
restriction of the vector bundle @xmath528 ( used in the construction of supermanifold @xmath187 in equation ( [ e : sjs ] ) ) on @xmath524 is holomorphic along @xmath2 .
additionally we can interpret component @xmath239 in the differential @xmath284 on @xmath328 as a contribution from a superextension of the cr structure defined in ( [ e : gfdjfur ] ) . from this
we deduce that the algebra @xmath529 we have constructed coincides with the cr tangential complex on an open subset of the manifold @xmath166 .
the integral form @xmath45 is the only cr - holomorphic form invariant with respect to @xmath530 . from this
we deduce that it coincides ( up to a multiplicative constant ) with the integral form defined by @xmath57 it is possible reconstruct an action of the super poincare group sp on @xmath72 .
the reader will find explanations of why the measure is not invariant with respect to the full superconformal group in section [ s : gufjasklly ] .
a tensor category of complex superspaces @xmath531 ( see ref.[dmiln ] for introduction to tensor categories ) has two real forms .
the first is a category of real superspaces @xmath532 .
it is more convenient to think about objects of this category as of complex superspaces , equipped with an antiholomorphic involution @xmath156 .
another tensor category related to @xmath531 is formed by complex superspaces , equipped with an antilinear map @xmath110 [ d : gdfjgdf ] suppose @xmath339 is a @xmath533-graded vector space over complex numbers .
an antilinear map @xmath534 is a graded real structure if @xmath535 we denote by @xmath536 a parity of @xmath537 . an element @xmath537 is real iff @xmath538 .
only even elements can be real with respect to a graded real structure .
a graded real superspace is a pair @xmath539 .
grader real superspaces form a tensor category @xmath163 we shall be mostly interested in categories @xmath531 and @xmath163 . the categories @xmath531 @xmath163 are related by tensor functors .
the first functor is complexification @xmath540 , @xmath541 .
it forgets about the map @xmath110 .
the second functor is @xmath542 , @xmath543 .
the object @xmath544 is a direct sum @xmath545 .
there is an antilinear isomorphism @xmath546 . for @xmath547
define @xmath548 by construction @xmath118 .
a language of tensor categories can be used as a foundation for developing commutative algebra and algebraic geometry .
if we start off in this direction with a category @xmath531 the result will turn to be algebraic supergeometry . a category @xmath163 provides us with some real form of this geometry .
first of all a @xmath163 or a real graded manifold is an algebraic supermanifold @xmath49 defined over @xmath108 .
the manifold @xmath49 carries some additional structure .
the manifold @xmath139 is equipped with an antiholomorphic involution @xmath549 .
there is also an antilinear isomorphism of sheaves of algebras @xmath550 such that @xmath118 .
there is a @xmath50 version of a @xmath163 manifold .
it basically mimics a stricture of @xmath50 completion of algebraic @xmath163 manifold at locus of @xmath549 fixed points .
any @xmath50 manifold admits a noncanonical splitting : @xmath551 , where @xmath552 is some complex vector bundle over @xmath139 .
a @xmath163-structure manifests in an antilinear automorphism @xmath110 of @xmath552 , that satisfies @xmath553 .
observe that @xmath110 , together with multiplication on @xmath132 defines a quaternionic structure on @xmath552 .
for any complex algebraic supermanifold @xmath49 there is the underlying @xmath163 manifold . as an algebraic supermanifold it is equal to @xmath554 .
there is a canonical antiinvolution on @xmath555 .
the morphism of sheaves ( [ e : jjshdga ] ) in local charts is defined by formulas similar to ( [ e : qdsfdfbshe ] ) .
this construction manifests itself in @xmath50-setting as follows .
any complex supermanifold @xmath49 defines a @xmath50 supermanifold @xmath556 that is holomorphic in odd directions .
it is a @xmath50-completion of @xmath557 near diagonal of @xmath558 .
suppose @xmath559 is a splitting of @xmath556 , where @xmath552 is a complex vector bundle on @xmath560 .
the vector bundle @xmath561 has a natural quaternionic structure and defines a @xmath50 @xmath163-manifold @xmath562 .
this manifold is isomorphic to completion of @xmath563 .
sometimes it is more convenient to work with the manifold @xmath564 .
suppose we are given an ordinary real lie group and a closed subgroup @xmath565 with lie algebras @xmath566 .
additionally we have a complex subgroup @xmath567 in complexification of @xmath317 , with lie algebras @xmath568 .
we assume that the map @xmath569 is a local embedding . by construction @xmath570
is a holomorphic homogeneous space .
it tangent space @xmath571 contain a subspace @xmath572 .
the operator @xmath53 is an operator of complex structure on @xmath570 . due to @xmath317-homogeneity spaces @xmath143 have constant rank and form a subbundle @xmath136 .
we can decompose @xmath573 .
it follows from the fact that @xmath175 is a subalgebra that the constructed distribution @xmath2 is integrable and defines a cr - structure . a condition that the map @xmath65 ( [ e : frtw ] ) is a local embedding is equivalent to @xmath574 . in other words @xmath575 it is easy to see that a fiber @xmath576 at a point @xmath212 is isomorphic to @xmath577 this construction of cr - structure can be extended to a supercase . a consistent way to derive such extension is to use a functorial language of ref . [ dm ] .
however since this exercise , which we leave to the interested reader , is quite straightforward we provide only the upshot .
we start off with description of a data that defines a homogeneous space of a supergroup .
a complex homogeneous space @xmath248 of a complex supergroup @xmath317 with lie algebra @xmath109 is encoded by : * 1.global data : * an isotropy subgroup @xmath578 ( closed , analytic , possibly nonconnected ) .
this data defines an ordinary homogeneous space @xmath579 ; * 2.local data : * a pair of complex super lie algebras @xmath580 such that @xmath581 . in the cases when we specify
only lie algebra of isotropy subgroup is clear from the context . a real graded structure on a homogeneous space @xmath582
is encoded by an antiholomorphic involution on @xmath583 that leaves subgroup @xmath584 invariant ; a graded real structure @xmath110 on @xmath109,such that @xmath585 .
if we are given a real subalgebra @xmath586 and a complex subalgebra @xmath587 such that @xmath588 we claim that a supermanifold @xmath589 carries a @xmath590-homogeneous cr - structure . in this section
we will discuss mostly general facts about cr structures specific to supergeometry .
suppose @xmath49 is a supermanifold equipped with an integrable cr distribution @xmath2 .
we present some basic examples of @xmath2-holomorphic vector bundles on @xmath49 .
* example * sections of vector bundle @xmath591 is a module over lie algebra of sections of @xmath592 .
thus the gluing cocycle of this bundle is cr - holomorphic .
it implies that the bundle @xmath593 is also cr - holomorphic .
suppose we have a trivial cr - structure on @xmath594 .
we assume that the space is equipped with global coordinates @xmath595 .
the algebra of tangential cr complex @xmath596 has topological generators @xmath597 .
denote by @xmath72 a subalgebra generated by @xmath598 and by @xmath599 a subalgebra generated by @xmath600 .
we have @xmath601 .
the algebra @xmath599 has trivial cohomology . as a result projection @xmath602 is a quasiisomorphism .
it turns out that this construction exists in a more general context of an arbitrary cr - manifold .
informally we may say think that a super - cr manifold is affine in holomorphic odd directions .
it parallels with the complex case .
the construction requires a choice of @xmath50-splitting of cr - manifold @xmath49 .
suppose @xmath259 is an ordinary manifold , @xmath552 is a vector bundle .
let @xmath559 denote the supermanifold whose sheaf of functions coincides with a sheaf of sections of the grassmann algebra of the bundle @xmath603 .
such supermanifold is called split . by construction
it admits projection @xmath604 .
any @xmath50 manifold is split , but the splitting is not unique .
a space of global functions on @xmath559 is isomorphic to a space of sections of the grassmann algebra @xmath605 of vector bundle @xmath552 . to make a connection with our considerations we identify @xmath606 .
denote @xmath607 restriction of @xmath608 on @xmath139 . in terms of the splitting operator @xmath4 can be encoded by a pair of operators of the first order @xmath609 . the lowest degree component in powers @xmath610 of the operator
@xmath611 is @xmath612 .
it is a @xmath613 linear map , with a locally free image .
the image @xmath614 of a splitting @xmath615 can be used to generate a differential ideal @xmath616 of @xmath617 .
denote the quotient @xmath618 by @xmath619 .
the complex @xmath619 is a differential graded algebra
. we can interpret it as a space of functions on some superspace @xmath198 .
a possibility to split @xmath620 implies smoothness of @xmath198 .
[ e : fdsafjh ] the map @xmath621 is a quasiisomorphism . follows from consideration of a spectral sequence associated with filtration @xmath622(we denote by @xmath623 the k - th power of the ideal generated by @xmath614 ) .
a cr - manifold @xmath49 is locally embeddable to @xmath624 if in a neighborhood of a point there is a collection of @xmath625 even and @xmath626 odd function that are annihilated by @xmath4 and whose jacobian is nondegenerate .
[ d : wpprnfj ] let us assume that @xmath627(superscript @xmath628 denotes the odd part),i.e .
dimension of the odd part of cr distribution is maximal possible .
we can locally generate ideal @xmath614 by elements @xmath629 and take @xmath15 as a @xmath4 closure of @xmath614 .
it is not hard to check that under such assumptions proposition [ e : fdsafjh ] holds .
denote by @xmath630 a submanifold specified by @xmath614 [ e : fdfdsdfv ] the lie algebra of infinitesimal automorphisms of cr - structure is equal to @xmath631\in { \overline{{\cal f } } } , \mbox { for all } b\in { \overline{{\cal f}}}\}\}$ ] with @xmath632 . in purely complex case
the quotient construction can be replaced by @xmath633\in { \overline{{\cal f } } } , \mbox { for all } b\in { \overline{{\cal f}}}\}\}$ ] and the extension @xmath634 has a splitting . by construction elements @xmath629 are invariant along vector fields from distribution @xmath2 .
we can guarantee that differential ideal generated by @xmath635 is invariant with respect to elements of @xmath636 . as a result we can push the action of @xmath636 to @xmath637-this
is familiar fact from super complex geometry .
this contrasts with absence of an action of @xmath638 or @xmath639 on the ideal @xmath15 and on @xmath637 for a general cr structure .
one can prove however that @xmath637 admits an a@xmath78 action of @xmath638 .
a partial remedy is to consider subalgebra @xmath640\in { \overline{{\cal f } } } , \mbox { for all } b\in { \overline{{\cal f}}}\}\}\subset out_{{{\cal f}}}$ ] .
this subalgebra acts upon @xmath637 .
however this algebra is trivial if the levi form of @xmath2 is not degenerate supergeometry provides us with a complex of cr - integral forms .
let @xmath641 be the super grassmann algebra of @xmath608 .
let @xmath642 be the berezinian line bundle of a real manifold @xmath49 .
denote @xmath643 the tensor product @xmath644 .
there is a pairing @xmath645 between sections @xmath646 and @xmath647 . the symbol @xmath18 denotes contraction of a differential form with a polyvector field .
the operation @xmath18 takes values in sections of @xmath642 .
the value @xmath648 can be used as an integrand for integration over @xmath49 . the orthogonal complement to the ideal @xmath649 is a subcomplex @xmath650 .
it is a differential graded module over @xmath651 .
[ p : fdfdhw ] let @xmath49 be a supermanifold with a cr distribution @xmath2 of dimension @xmath652 .
there is an isomorphism @xmath653 compatible with a stricture of @xmath637-module .
the isomorphism is unique .
using @xmath50 splitting we can identify @xmath654 and @xmath655 with sections of some vector bundles @xmath656 and @xmath657 over @xmath139 .
it is fairly straightforward to establish an isomorphism of @xmath656 and @xmath657 with the help of the splitting .
in particular there is a @xmath50 isomorphism @xmath658 .
one can think about @xmath659 as of a space of functions on a supermanifold @xmath630 .
then a space of sections @xmath660 coincide with @xmath661 .
it elements can be integrated over @xmath630 .
the integral defines a pairing @xmath662 , which is nondegenerate .
an element @xmath663 defines a functional @xmath664 ( @xmath665 )
. we may think about it as of an integral @xmath666 , where @xmath667 .
such interpretation of the integral uniquely specifies map @xmath132 . since @xmath668 is invertible the induced isomorphism @xmath669 is compatible with @xmath4 ( use pairings @xmath670 to check this ) .
suppose that a super cr - manifold @xmath49 is cr embedded into a complex super manifold @xmath140 .
denote by @xmath53 an operator of complex structure in tangent bundle @xmath671 .
we assume @xmath672 .
denote by @xmath673 the complex berezinian of @xmath140 .
an easy local computation shows that @xmath674 is isomorphic to @xmath593 .
suppose @xmath140 is a calabi - yau manifold , i.e. it admits a global nonvanishing section @xmath57 of @xmath673 .
a restriction of @xmath57 on @xmath49 defined a global cr - holomorphic section of @xmath593 .
an isomorphism of proposition [ p : fdfdhw ] provides a @xmath4 closed section of @xmath675 . the proof proposition [ p : fdfdhw ] parallels with the proof of serre duality in super case given in ref .
haske and wells used sheaf - theoretic description of a complex supermanifold , which significant simplify the argument .
the main simplification comes from the local poincare lemma , which is absent is cr - case .
it is worthwhile to mention that there is no canonical ( @xmath638 or @xmath639 equivariant ) map @xmath676 $ ] .
this seems to be one of fundamental distinctions of purely even and super cases .
we think the reason is that the only known construction of this map is through the intermediate complex @xmath654 . as have
have already mentioned in remark ( [ e : fdfdsdfv ] ) the complex @xmath637 carries only a@xmath78 action of @xmath638 .
a real line bundle @xmath642 on a complex super manifold is a tensor product @xmath677 , where @xmath678 is a holomorphic berezinian .
a decomposition @xmath679 implies that @xmath680 [ e : hhghssa ] suppose @xmath49 is a complex @xmath66-dimensional manifold , @xmath552 is an @xmath681-dimensional vector bundle . on the total space @xmath682
there is a canonical section of @xmath683 .
in local odd fiberwise coordinates @xmath684 it is equal to @xmath685 [ p : kahedy ] the forms @xmath686 and @xmath687 are @xmath4-closed . since the formula ( [ e : jsdfdmdas ] )
does not depend on a choice of coordinates on @xmath49 one can do a local computation , which is trivial .
the author would like to thank p. candelas , p.deligne , m.kontsevich , l. maison , a.s .
schwarz for useful comments and discussions .
the note was written while the author was staying at mittag - leffler institut , max plank institute for mathematics and institute for advanced study .
the author is grateful to these institutions for kind hospitality and support .
after the preprint version of this paper [ mov ] had been published , the author learned about the work by l.j.mason , d.skinner [ ms ] where they treat a similar problem .
[ klr73 ] m. f. atiyah , n. j. hitchin , v. g. drinfeld , yu . i. manin , _ construction of instantons _ phys .
lett . a 65 ( 1978 ) , no .
3 , 185187 .
n. berkovits _ covariant quantization of the superparticle using pure spinors _ jhep 0109 ( 2001 ) 016 , hep - th/0105050 p. deligne , j. s. milne , _ tannakian categories , in
hodge cycles , motives and shimura varieties _ , lnm 900 , 101 - 228 .
p. deligne , j.w .
notes on supersymmetry ( following joseph bernstein ) .
quantum fields and strings : a course for mathematicians _ , vol . 1 , 2 ( princeton , nj , 1996/1997 ) , 4197 , amer .
soc . , providence , ri , 1999 .
j. frauendiener , r. penrose _ twistors and general relativity .
_ mathematics unlimited2001 and beyond , 479505 , springer , berlin , 2001 ph .
griffiths , j. harris _ principles of algebraic geometry . _
pure and applied mathematics .
wiley - interscience , john wiley & sons , new york , 1978 .
xii+813 pp . c. haske ,
r. o. wells , jr._serre duality on complex supermanifolds _ duke math .
j. 54 , no . 2 ( 1987 ) , 493500 n. ishibashi , h. kawai , i. kitazawa , and a. tsuchiya _ a large - n reduced model as superstring _
, nucl . phys . *
b492 * ( 1997 ) kadeishvili_the algebraic structure in the homology of an a@xmath78-algebra _ soobshch .
nauk gruzin .
ssr , 1982 yu.i .
manin _ gauge field theory and complex geometry_grundlehren der mathematischen wissenschaften , 289 .
springer - verlag , berlin , 1997 .
m. markl _ transferring a@xmath78 ( strongly homotopy associative ) structures_math.at/0401007 l.j.mason , d.skinner _ an ambitwistor yang - mills lagrangian _ phys.lett .
b636 ( 2006 ) 60 - 67 , hep - th/0510262 m. movshev _ yang - mills theory and a superquadric _ hep - th/0411111 m. movshev , a. schwarz _ on maximally supersymmetric yang - mills theories .
_ , nuclear physics b , volume 681 , issue 3 , p. 324 - 350
m. movshev , a. schwarz _ algebraic structure of yang - mills theory _ the unity of mathematics , 473523 , progr .
, 244 , birkhuser boston , boston , ma , 2006 , hep - th/0404183 a. schwarz _ semiclassical approximation in batalin - vilkovisky formalism_commun.math.phys . 158 ( 1993 ) 373 - 396 , hep - th/9210115 a. schwarz _ topological quantum field theories _ hep - th/0011260 v. v. serganova _ classification of simple real lie superalgebras and symmetric superspaces _
( russian ) funktsional .
i prilozhen .
17 ( 1983 ) , no .
3 , 4654 f.pellegrini _ on real forms of complex lie superalgebras and complex algebraic supergroups_math.ra/0311240 r. penrose _ twistor theory after 25 years its physical status and prospects .
twistors in mathematics and physics _ london math .
lecture note ser .
, 156 , cambridge univ . press , cambridge , 1990 .
e. witten _ an interpretation of classical yang - mills theory _
preprint hutp-78/a009 e. witten _ gauge theory as a string theory in twistor space _ hep - th/0312171 | we construct a real - analytic cr supermanifold @xmath0 , holomorphically embedded into a superquadric @xmath1 .
a cr distribution @xmath2 on @xmath0 enables us to define a tangential cr complex @xmath3 .
we define a @xmath4-closed trace functional @xmath5 and conjecture that a chern - simons theory associated with a triple @xmath6 is equivalent to n=3 , d=4 yang - mills theory with a gauge group u@xmath7 .
we give some evidences to this conjecture .
.2 in _ dedicated to yu.i .
manin on his 70-th birthday _ .2 in |
in their first formulation of quantum mechanics , both schrdinger and dirac were strongly influenced by connections to the hamiltonian formulation of classical mechanics .
indeed , in one of the very first applications of schrdinger s time - dependent equation ( tdse ) , dirac @xcite indicated a close parallel between the coupled first - order set of equations arising from the tdse and the coupled first - order hamilton equations of classical mechanics .
dirac introduced the time - dependent basis set expansion , @xmath1 where the @xmath2 are complex co - efficients and @xmath3 denotes an arbitrary basis . in the tdse ( where , for the moment , we put @xmath4 ) this expansion leads to the set of first - order coupled equations , @xmath5 where @xmath6 are the matrix elements of the quantum hamiltonian .
dirac then remarked that by considering the co - efficients @xmath7 and @xmath8 to be canonical variables and assuming a hamiltonian function @xmath9 the quantum equations are equivalent to the classical hamilton equations @xmath10 note , however , that this is still fully quantum mechanical since the matrix elements and the amplitudes appearing in the classical hamiltonian are all complex objects .
hence , there is no obvious classical counterpart .
dirac also made a transformation to real variables by using amplitude and phase of the quantum co - efficients , i.e.@xmath11 where @xmath12 .
although with this transformation the variables are real , the classical hamiltonian still contains complex quantities and hence does not obviously correspond to any real physical system .
forty years later , strocchi @xcite approached the question of quantum / classical equivalence slightly differently .
he first formulated classical mechanics in terms of complex variables @xmath13 , with @xmath14 and @xmath15 real , to give the hamiltonian equations ( [ eq : hameqns ] ) in the form , @xmath16 and its complex conjugate . here
@xmath17 _ is _ a real classical hamiltonian .
then it is remarked that the tdse coupled equations ( [ eq : schreqns ] ) are equivalent to these classical equations i.e. @xmath18 if the classical hamiltonian function is taken as the expectation value of the quantum hamiltonian i.e. @xmath19 as in the treatment of dirac .
although ostensibly time - dependent and complex , it is easy to prove that @xmath17 becomes constant and real for hermitian hamiltonians . identifying @xmath2 with @xmath20 and reverting to the real variables @xmath21 one has however , still the apparently complex form quoted by strocchi @xcite , @xmath22 of course , for @xmath23 hermitian one does achieve a real classical hamiltonian @xmath24\ ] ] and in the special case that all coupling matrix elements are real the hamiltonian @xmath25 which is that of coupled real harmonic oscillators .
note that the coupling is of a very special form in which there is both linear position and momentum off - diagonal coupling with exactly the same coupling strengths .
the mapping of the tdse onto classical - like equations up until now has been regarded as something of a curiosity , although it has been used as a starting point to treat some problems of molecular electronic dynamics semi - classically @xcite .
here we show that systems of coupled classical oscillators corresponding to the hamiltonian eq .
( [ eq : classham ] ) can be realised whose position and momentum variables reproduce _ exactly _ the time - dependent coefficients of an expansion of the quantum wavefunction .
we call these the p&q - coupled oscillators .
each oscillator plays the role of a state in hilbert space and the coupling between the oscillators the role of quantum coupling matrix elements .
unfortunately , for a general quantum system the corresponding exactly equivalent classical system involves rather complicated coupling schemes between the oscillators .
however , we show that in a weak - coupling approximation , simpler sets of coupled oscillators in which only the q - coupling in the hamiltonian eq .
( [ eq : classham ] ) occurs provide an exceedingly good approximation to the exact result . in previous papers @xcite , where we studied energy transfer along dipole - dipole interacting molecules , we called this approximation the realistic - coupling approximation ( rca ) since for this case , either classical or quantum , to be realistic it is essential that the oscillators and quantum entities largely retain their character when coupled i.e. the coupling is weak compared to their internal forces .
although in this work we consider a general hamiltonian where the diagonal elements in a chosen basis are not necessarily associated with concrete physical entities and thus there is no _ a priori _ reason for the off - diagonal elements to be small , we will still keep the term rca for this weak coupling approximation .
it is interesting that , with reference to specific systems , several authors have already recognised the similarity between q - coupled classical oscillators and few - level quantum systems . indeed , already in one of the the first quantum treatments of resonant electronic energy transfer between identical atoms , frenkel in 1930 @xcite remarked on the essential equivalence of the quantum treatment to the earlier 1925 classical treatment of holtsmark @xcite .
in particular he showed that the normal mode frequencies of the classical coupled equations reduce to the quantum eigenvalues in the approximation that we call the rca . in 1977
mckibben @xcite constructed a system of three coupled mechanical pendula and showed that , in a weak - coupling approximation similar to our rca , their equations of motion become equivalent to the quantum equations describing the operation of a spin filter on three levels of the hydrogen atom .
later hemmer and prentiss @xcite used the same 3-pendulum classical system to interpret the quite different 3-level atomic problem of the stimulated resonance raman effect and pointed out `` the strong mathematical similarities between the pendulum amplitude equations and the resonance raman equations '' .
more recently , marx and glaser @xcite have derived an exact correspondence ( in the rca ) between the dynamics of three isotropically coupled spins and those of three coupled classical oscillators . here
the liouville equation for the density matrix and spin correlation functions derived from it were considered . in a rather different application ,
jolk et al .
@xcite and kovaleva et al .
@xcite have pointed out the similarity of the non - crossing behaviour of the eigenfrequencies of a pair of dissimilar classical oscillators under varying coupling to the non - crossing behaviour of the eigenenergies of two quantum levels subject to a varying perturbation , as in the celebrated landau - zener problem .
similarly spreeuw et al .
@xcite has discussed a classical optical system as analogue to a driven two - level atom .
in previous communications @xcite we have shown that the time - dependent schrdinger equation for an aggregate of @xmath0 coupled monomers having a single electronic transition and the classical equations for coupled electric dipoles are equivalent in the rca .
there we demonstrated that the quantum coherence in the transfer of electronic excitation along a linear chain of monomers is reproduced by the transfer of electric dipole strength along the equivalent classical array .
the equivalence proof was restricted to this special problem of coupled classical electric dipoles and its relevance for energy transfer in photosynthesis .
these many and varied applications point to a general property of equivalence of classical oscillators and quantum systems in finite hilbert spaces .
however , none of the above authors discuss their findings in the context of dirac and strocchi s analysis .
also they all use classical oscillators that are q - coupled , i.e. they work only in the rca and do not recognise that it is possible to construct classical oscillators that _ exactly _ mimic the quantum system . in the present paper
we show how to construct such p&q - coupled oscillators and then clarify under what conditions the rca allows simpler q - coupled sets of oscillators to be used as analogues of coupled - state quantum systems .
the structure of the paper is as follows . in sec.2.1
we derive the classical p&q - coupled equations whose variables give directly the coefficients of the time - dependent schrdinger wavefunction .
we also examine the derivation of the rca and the meaning of this approximation in the context of the dirac and strocchi mapping of the schrdinger equation to hamilton s equations . as specific example , in section 3 we consider excitation transfer on a quantum aggregate of @xmath0 coupled monomers and its classical equivalent of @xmath0 coupled pendula .
special attention is given to the @xmath26 dimer case whose wavefunction is considered a fundamental example of quantum entanglement . in this case
the quantum ( and therefore also the p&q - coupled classical ) solution can be obtained in simple closed from
. then the rca is tested by comparing the q - coupled classical solution with the p&q - coupled solution i.e. with the _ exact _ quantum solution .
oscillatory energy transfer on the dimer is shown to correspond classically to the well - known `` beating '' phenomenon of coupled pendula from whose motion the entangled wavefunction can be extracted .
the classical analysis is given , as example , for the specific case of coupled undamped vertical pendula .
however it is clear that the analysis also applies to other sets of oscillators , particularly classical electrical lc circuits , as shown in appendix [ elec_osci ] .
it is also demonstrated in appendix [ damping ] that the inclusion of a velocity - dependent damping term in the classical equations corresponds exactly to the inclusion of damping via a complex eigenvalue in the quantum equations , as used for example in ref .
first we will show that the first - order schrdinger equations ( [ eq : schreqns ] ) map simply onto the classical hamilton equations for a set of coupled classical oscillators .
since the classical problem of coupled oscillators is almost always formulated using the newton equations rather than the hamilton equations we also present the corresponding second - order newton equations .
we start with the strocchi classical hamiltonian , eq.([eq : classham ] ) . the hamilton equations ( [ eq : hameqns ] ) give @xmath27 symbolically , writing @xmath28 and @xmath29 as vectors and @xmath23 as a matrix the above equations are @xmath30 and formally @xmath31 which are a set of coupled oscillator equations and can be solved for @xmath32 and @xmath33 .
the momenta at time @xmath34 can then be calculated from @xmath35 the set of complex amplitudes , the vector @xmath36 , is constructed as @xmath37 from the hamilton eqs .
( [ eq : hameq ] ) we have @xmath38 similarly the schrdinger equation ( [ eq : schreqns ] ) is written @xmath39 or @xmath40 which is exactly the classical equation ( [ eq : zclass ] ) . hence the p&q - coupled classical equations and the coupled quantum schrdinger equations are identical and in particular they have the same eigenvalues .
to see how to construct a set of real classical oscillators , now we write the formal solution in terms of the individual oscillator amplitudes .
first we split off the diagonal term in the hamiltonian .
this quantum energy @xmath41 ( divided by @xmath42 ) will be denoted by @xmath43 .
the off - diagonal elements ( divided by @xmath42 ) will be denoted by @xmath44 .
then the hamilton equations give @xmath45 forming @xmath46 gives @xmath47 which are identical to the quantum eqs .
( [ eq : schreqns ] ) by construction . taking the time derivative of the @xmath48 equation and substituting for the @xmath49 from the second equation leads to the coupled second - order equations @xmath50 q_m\ ] ] where we defined @xmath51 this set of classical equations are the p&q - coupled equations .
although the coupling elements appear quite complicated , for a given set of quantum matrix elements @xmath52 they can , in principle , be solved to obtain @xmath53 and @xmath54 .
then the time - dependent momenta must be calculated by matrix inversion as in eq .
( [ eq : pdef ] ) and the complex @xmath55 , equal to the quantum coefficients @xmath56 , calculated . to construct a classical analogue to a given quantum system
the oscillators must be coupled as in eq .
( [ eq : coupleqs ] ) .
however , when the hamiltonian involves only positional @xmath28 couplings , the hamilton equations simplify to @xmath57 or equivalently @xmath58 which are the q - coupled classical equations .
if we define real coupling elements @xmath59 these are the standard coupled equations , for example of a set of linearly - coupled mechanical or capacitatively -coupled electrical oscillators . in appendix a it is shown that in the rca , which corresponds to having @xmath60 , the solutions of the simple q - coupled equations are a good approximation to the more complicated p&q - coupled equations ( [ eq : coupleqs ] ) + in the basis @xmath3 defined in eq .
( [ eq : basis_expansion ] ) the hamiltonian defined in ( [ eq : schreqns ] ) is not diagonal in the general case . to find the eigenfunctions and eigenenergies we solve @xmath61 the eigenstates @xmath62 can be expressed in the original basis via @xmath63 denoting the diagonal elements @xmath64 and the off diagonal elements by @xmath65 the co - efficients @xmath66 can be obtained from the coupled set of equations @xmath67 a general time - dependent wavefunction ( coherent wavepacket ) can be expanded either in the basis @xmath3 or in the eigenbasis @xmath62 , i.e. @xmath68 since the eigenbasis diagonalises the hamiltonian , the coefficients @xmath69 are given simply by , @xmath70 where the time - independent complex coefficients @xmath71 are decided by the initial conditions . then using eq .
( [ eq : k - to - n ] ) in eq .
( [ eq : psit ] ) one has the amplitude of state @xmath72 , @xmath73 the newton equations of motion for the exactly equivalent set of classical oscillators are the coupled equations eqs .
( [ eq : coupleqs ] ) .
the eigenfrequencies of these classical equations are identical to the @xmath74 of the diagonalised quantum problem .
since the eigenvalues and therefore the eigenfunctions of the quantum and classical systems are identical , we see that the classical time dependence can be used to construct the quantum wavefunction .
the normal modes arising from the diagonalization of eqs .
( [ eq : coupleqs ] ) are of the form @xmath75 where @xmath76 and @xmath77 are real constants . from these normal modes
we derive the velocities @xmath78 then the complex classical amplitudes are @xmath79 to obtain the general solution as a sum of normal modes , @xmath80 where we have to set the @xmath81 initial conditions so that @xmath82 and put @xmath83 . comparing this result with eq .
( [ eq : quantcnt ] ) we see that the classical amplitudes @xmath55 reproduce the quantum co - efficients @xmath56 and the classical motion can be used to reconstruct the quantum entangled wavefunction of eq .
( [ eq : psit ] ) . as we show in appendix a , in the rca the eqs .
( [ eq : coupleqs ] ) reduce to become identical in form to the simpler linearly @xmath28-coupled classical equations ( [ eq : st.class ] ) , which we write , @xmath84 now we show that , again in rca , these equations have eigenfrequencies which closely approximate those of the quantum problem and therefore also the eigenfrequencies of eqs .
( [ eq : coupleqs ] ) .
substitution of the special eigenmode solution @xmath85 gives @xmath86 this equation can be written @xmath87 now we make a second time the realistic coupling approximation ( rca ) , by considering that the spread of eigenfrequencies ( bandwidth of the dispersion relation ) resulting from diagonalisation of eq.([eq : ticlass ] ) is small compared to the mean natural frequency @xmath88 of the oscillators . similarly , for non - identical oscillators , the width of the @xmath43 distribution must be small compared to @xmath89 . then on the r.h.s . of eq .
( [ eq : ticlass ] ) we can approximate @xmath90 and @xmath43 by @xmath89 to give , @xmath91 further , if we make the identification @xmath92 and @xmath93 , eqs .
( [ eq : tiquant ] ) and ( [ eq : tiquantclass ] ) are identical and we can put in rca @xmath94 note that the rca is synonymous with the condition that the couplings @xmath95 are small compared to @xmath89 and equivalently in the quantum case that all @xmath96 , are small compared to the mean energy @xmath97 of the various eigenvalues @xmath98 of @xmath99 .
note also that this in no way implies that perturbation theory must be applicable since , as in the examples given below , the analysis applies to the case where all @xmath98 are degenerate , where perturbation theory is invalid .
to illustrate the construction of classical oscillator systems which can reproduce entangled wavefunctions , both exactly and approximately in rca , we consider the quantum problem of @xmath0 identical two - level systems .
specifically we consider the dynamics when there is exactly one excitation present and take @xmath3 as the state in which monomer @xmath72 is excited and all other monomers are in the ground state .
this is the exciton model studied originally by frenkel .
such model quantum systems are of fundamental importance in several areas , for example in quantum computing and in the modelling of the photosynthetic unit , in addition to applications mentioned in the introduction .
indeed , the quantum dimer of @xmath26 coupled two - level systems ( two qubits ) , is viewed as having a wavefunction which is the simplest example of quantum entanglement .
this is the case we discuss first . to make the problem concrete
we will think of a dimer composed of two identical atoms or molecules each having only a ground and one excited state .
note , that we restrict to the subspace where exactly one excitation is present .
classically this corresponds to two coupled oscillators .
the dimer has @xmath100 and @xmath101 eigenstates of the form , @xmath102 with eigenenergies @xmath103 , where @xmath97 is the monomer transition energy and @xmath104 is real .
the state @xmath105 has monomer @xmath106 excited and monomer @xmath107 in the ground state and correspondingly for @xmath108 .
we expand a solution of the tdse as @xmath109 the two eigenstates propagate independently in time according to , @xmath110 = a_\pm \exp[-(i/\hbar ) ( \epsilon\pm \mathcal{v})t]\ ] ] excitation transfer is described by the initial condition @xmath111 which leads to the time - dependence , @xmath112 with , @xmath113 .
$ ] these expressions can be simplified further to give @xmath114 \cos [ \mathcal{v}t/\hbar]\\ c_2(t ) = & - i \exp[-(i/\hbar ) \epsilon t ] \sin [ \mathcal{v}t/\hbar]\nonumber\end{aligned}\ ] ] which are the exact quantum solutions and describe a periodic transfer of excitation between the two monomers .
note that in forming the density matrix of coefficients @xmath115 the pure phase factor @xmath116 $ ] disappears . the mapped hamilton equations , with @xmath117 and @xmath118 , for the case @xmath26 are @xmath119 to give the coupled oscillator equations @xmath120 in the usual way these symmetric equations can be diagonalised by the transformation @xmath121 to give normal modes @xmath122 with eigenfrequencies @xmath123 , where we take the positive square root and assume @xmath124 .
as they should , these reproduce exactly the eigenenergies @xmath103 of the quantum dimer problem .
the momenta are obtained from the hamilton equations as @xmath125 which combine conveniently to give @xmath126 thus in the eigenmodes we have the simple expression for the quantum amplitudes in terms of the classical amplitudes @xmath127 note that all the above results are invariant to a change of sign of @xmath128 which only serves to flip the @xmath129 eigenvalues .
as example , in the beating mode the wavefunction is exactly reproduced by the classical amplitudes .
the initial conditions for beating are @xmath130 leading to amplitudes @xmath131 \end{split}\ ] ] forming the velocities and from them the momenta , after some algebra one finds @xmath132 $ ] so that @xmath133 .
\end{split}\ ] ] choosing @xmath134 and noting that @xmath135 , these are exactly the quantum amplitudes @xmath136 of eq .
( [ eq : psit ] ) leading to the explicit beating forms of eqs .
( [ eq : c1quant ] ) .
we now show how to simulate the quantum equation of motion using two coupled classical pendula .
the classical equations of motion for two ( mathematical , i.e linearized ) pendula with oscillation angle @xmath137 , natural frequency @xmath89 and coupled by a spring with coupling strength @xmath138 are @xmath139 making again the identification @xmath140 the equations ( [ eq : exmap ] ) which map exactly the schrdinger equation , can be written in the same form @xmath141 where @xmath142 so that in this case @xmath143
. thus , one takes two pendula of natural frequency @xmath89 .
one then couples them with strength @xmath138 and simultaneously adjusts the lengths to give a new natural frequency @xmath144 .
then the oscillation amplitudes and velocities of this classical system reproduce _ exactly _ the complex time - dependent amplitudes of the quantum dimer wavefunction , with transition energy @xmath145 and coupling matrix element @xmath140 .
clearly , as alternative one can leave the natural frequency unchanged as @xmath144 and then , for given @xmath138 infer the transition energy @xmath89 of the equivalent quantum system from @xmath144 .
hence we have shown how to construct a pair of classical oscillators whose motions reproduce the entangled two - qubit quantum wavefunction time - dependence .
previous works pointing out the equivalence of classical oscillator motion and quantum time - development have not used the exact mapping but rather the simpler standard equations ( [ eq : phieqns ] ) or their equivalent and then invoked the rca .
next we investigate the accuracy of this approach in the simple dimer case .
the eigenfrequencies of the exact mapping eqs .
( [ eq : st.osc ] ) are readily calculated to be @xmath146 and @xmath147 . with @xmath148 this translates to the eigenfrequencies @xmath149 of eqs .
( [ eq : exmap ] ) and of course of the quantum problem .
the eigenfrequencies of the standard eqs .
( [ eq : phieqns ] ) are @xmath150 and @xmath151 involving the natural oscillator frequency .
however in the realistic coupling approximation ( rca ) we expand , as in section [ sec : eigenfunc+modes ] , @xmath152 to give eigenfrequency difference @xmath153 , as in the exact mapping .
this implies that , when the rca is valid , the solutions to the standard eqs .
( [ eq : phieqns ] ) will be a good approximation to the exact solution , up to an overall phase caused by the shift in absolute value of the eigenfrequencies .
this explains the previously - observed close agreement of quantum solutions and standard ( q - coupled ) oscillator amplitudes .
[ t]@xmath154 [ t]@xmath155 [ t]@xmath156 [ t]@xmath157 [ t]@xmath158 [ t]@xmath159 [ t]@xmath160 [ t]@xmath161 time dependence of the quantum or p&q - coupled classical ( blue ) and q - coupled classical ( red , dashed ) motion .
the left column shows details for short times and the right column is for longer times .
the upper row shows the absolute value squared and the middle row the real part of the amplitudes .
the bottom panel shows @xmath162 and @xmath163 .
the time is given in units of @xmath164 or equivalently @xmath165,title="fig : " ] [ t]@xmath161 time dependence of the quantum or p&q - coupled classical ( blue ) and q - coupled classical ( red , dashed ) motion .
the left column shows details for short times and the right column is for longer times .
the upper row shows the absolute value squared and the middle row the real part of the amplitudes .
the bottom panel shows @xmath162 and @xmath163 .
the time is given in units of @xmath164 or equivalently @xmath165,title="fig : " ] + [ t]@xmath156 time dependence of the quantum or p&q - coupled classical ( blue ) and q - coupled classical ( red , dashed ) motion .
the left column shows details for short times and the right column is for longer times .
the upper row shows the absolute value squared and the middle row the real part of the amplitudes .
the bottom panel shows @xmath162 and @xmath163 .
the time is given in units of @xmath164 or equivalently @xmath165,title="fig : " ] [ t]@xmath161 time dependence of the quantum or p&q - coupled classical ( blue ) and q - coupled classical ( red , dashed ) motion .
the left column shows details for short times and the right column is for longer times .
the upper row shows the absolute value squared and the middle row the real part of the amplitudes .
the bottom panel shows @xmath162 and @xmath163 .
the time is given in units of @xmath164 or equivalently @xmath165,title="fig : " ] + [ t]@xmath156 [ t ] time in @xmath164 time dependence of the quantum or p&q - coupled classical ( blue ) and q - coupled classical ( red , dashed ) motion .
the left column shows details for short times and the right column is for longer times .
the upper row shows the absolute value squared and the middle row the real part of the amplitudes .
the bottom panel shows @xmath162 and @xmath163 .
the time is given in units of @xmath164 or equivalently @xmath165,title="fig : " ] [ t]@xmath161 time dependence of the quantum or p&q - coupled classical ( blue ) and q - coupled classical ( red , dashed ) motion .
the left column shows details for short times and the right column is for longer times .
the upper row shows the absolute value squared and the middle row the real part of the amplitudes .
the bottom panel shows @xmath162 and @xmath163 .
the time is given in units of @xmath164 or equivalently @xmath165,title="fig : " ] + .
[ t]@xmath154 [ t]@xmath155 [ t]@xmath156 [ t]@xmath157 [ t]@xmath158 [ t]@xmath159 [ t]@xmath160 [ t]@xmath161 same as fig.1 except that the coupling strength is @xmath166 . ,
title="fig : " ] [ t]@xmath161 same as fig.1 except that the coupling strength is @xmath166 .
, title="fig : " ] + [ t]@xmath156 same as fig.1 except that the coupling strength is @xmath166 . ,
title="fig : " ] [ t]@xmath161 same as fig.1 except that the coupling strength is @xmath166 .
, title="fig : " ] + [ t]@xmath156 [ t ] time in @xmath164 same as fig.1 except that the coupling strength is @xmath166 . ,
title="fig : " ] [ t]@xmath161 same as fig.1 except that the coupling strength is @xmath166 .
, title="fig : " ] + to assess the accuracy of the rca we have calculated exact quantum ( exactly equivalent to the solution of the p&q - coupled classical eqs .
( [ eq : exmap ] ) ) and compared to solutions of the standard q - coupled classical eqs .
( [ eq : phieqns ] ) for the @xmath26 dimer case with `` beating '' initial conditions .
the results are shown in figs .
[ fig : classical ] and [ fig : classical2 ] . in fig .
[ fig : classical ] the case @xmath167 is shown and one sees excellent agreement between exact , p&q- coupled ( blue , solid ) and standard q - coupled ( red , dashed ) results over several periods of the transfer time ( the time is given in units of @xmath164 or equivalently @xmath165 ) . only a small shift in relative phase is perceptible . in the middle column
the q - coupled results are multiplied by an overall phase factor @xmath168 to compensate for the overall energy shift between classical and quantum eigenvalues .
the upper figures show the absolute values squared of the classical q - coupled and quantum coefficients where this phase difference disappears .
similarly the lower panel shows the imaginary part of the off - diagonal density matrix elements @xmath162 and @xmath163 where the overall phase factor cancels also ( the real part is zero ) .
by contrast to the good agreement for @xmath169 , the case @xmath170 is shown in fig .
[ fig : classical2 ] , where one sees the beginning of the breakdown of the rca in that , although still of the same shape , the phase difference between quantum and classical q - coupled curves is becoming more pronounced .
for @xmath26 we have shown that by adjusting the natural oscillator frequency to be @xmath171 one can readily construct a classical system to mimic exactly the quantum wavefunction . to illustrate the complexity of the equivalent classical system that begins to arise for quantum systems of larger hilbert space dimension , we consider next the extension to @xmath0 interacting quantum monomers .
this is a model for frenkel exciton energy transfer in molecular crystals @xcite , on dye aggregates @xcite or in the photosynthetic unit @xcite . in a previous paper @xcite
we have shown that in the rca a set of q - coupled classical oscillators gives energy propagation characteristics indistinguishable from the exact quantum result . for simplicity
we take the monomers to be identical with transition energy @xmath97 .
furthermore we take only nearest - neighbour interaction into account and set @xmath172 . for circular boundary conditions
the result of diagonalisation is standard @xcite and gives @xmath173 with @xmath174 .
the transformation matrix elements are @xmath175 if we consider the time development beginning with only monomer @xmath176 excited , i.e. @xmath177 , then from the orthogonality of normal modes one has @xmath178
the exact mapping of this quantum system to the hamilton equations gives the p&q - coupled eqs .
( [ eq : coupleqs ] ) in the form @xmath179 note that , although the first - order schrdinger equations involve nearest - neighbour couplings , the equivalent second - order newton equations explicitly contain next - nearest neighbour couplings . substituting the trial solution
@xmath180 gives the eigenfrequency equation @xmath181 with solutions @xmath182 in agreement with the quantum result eq .
( [ eq : dispersequant ] ) .
hence for this @xmath0-monomer case we see again how to construct a system of classical oscillators to reproduce _ exactly _ the quantum results .
however , the couplings between oscillators is becoming more complicated .
the simpler q - coupled classical equations are those with nearest - neighbour coupling only , and were solved numerically in @xcite to compare with the quantum energy transfer result . here
we show analytically that in rca one obtains the same eigenfrequencies and time - dependence as in quantum or p&q - coupled classical case .
we consider a ring of identical coupled pendula .
the coupling is between adjacent pendula only .
the standard equations of motion are , @xmath183 leading to the eigenmode equation @xmath184 b_{kn } = -k ( b_{k , n+1 } + b_{k , n-1}).\ ] ] the eigenvalues are @xmath185 which , in the rca becomes @xmath186 now we make the identification @xmath187 and @xmath188 to give , @xmath189 comparison to the quantum dispersion relation eq .
( [ eq : dispersequant ] ) shows an overall shift of eigenenergies by @xmath190 .
this is immaterial as it leads only to an overall phase factor .
hence , as was shown numerically in @xcite for a particular set of initial conditions , the q - coupled classical oscillators describe to vey good approximation the exciton dynamics on an @xmath0-monomer chain .
we have shown how the eigenenergies and eigenfunctions of an @xmath0-level quantum system with real coupling matrix elements can be reproduced by a suitable array of coupled classical oscillators .
hence , by observing the classical motion , one can reconstruct the quantum mechanical time - dependent wavefunction or density matrix .
the difficulty of realisation of a classical system mimicking the quantum entanglement , is mainly in the condition that a realistic system must reproduce elements of the quantum couplings faithfully .
this will be a task of increasing difficulty as the complexity of the quantum system increases .
nevertheless , we have shown explicitly that such a parallel is feasible for a circular or linear array of identical monomers .
this is realised by a corresponding array of coupled pendula .
similarly , the excitation transfer on a molecular dimer , or similar coupled two - level quantum systems , can be faithfully simulated by the beating motion of a pair of classical oscillators , either mechanical or electrical .
such experiments are carried out routinely in undergraduate physics laboratories .
realistically the multi - oscillator couplings necessary to construct p&q - coupled oscillators could be attained more easily with miniature lc oscillators than with mechanical ones .
however , we have shown also that the complexity of p&q - coupled oscillators can be circumvented by using only q - coupled oscillators in the rca .
this explains the success of previous comparisons of quantum time - dependent motion with that of physical q - coupled oscillator systems . throughout
we have neglected the effects of coupling of the quantum system or the classical oscillators to the environment .
of course in real systems such de - cohering or dissipating couplings are omnipresent . in the quantum case in the form of radiative or non - radiative decay of excited states or in eigenenergy shifts . in the classical case in the form of frictional forces for pendula or the equivalent resistance in the electrical case . at the phenomenological level of simply assigning a complex energy to quantum levels or , equivalently , adding a term proportional to velocity to the classical equation of motion , it is easy to show that the quantum / classical equivalence is preserved ( see appendix [ damping ] ) .
we have undertaken a more detailed study of this equivalence in the framework of standard theories of open quantum systems @xcite .
similarly , the correspondence between inducing radiative transitions in quantum systems and driving classical oscillators with external fields is under investigation .
we are grateful to prof .
hanspeter helm for many helpful discussions and assistance .
the observation that the rca is equivalent to the rotating - wave approximation is due to prof .
gerhard stock and we are grateful for this insight .
in the foregoing we have derived the rca by approximating the classical eigenmode frequencies .
further light can be shed on the nature of this approximation by looking at alternative derivations .
a useful variation is to include explicitly the eigenvalue phases in the quantum time development of eq .
( [ eq : psit ] ) i.e. to write @xmath191 forming the equivalent classical hamiltonian as before we obtain @xmath192 where we have set @xmath193 .
note that if one proceeds to second quantization and elevates the coefficients @xmath194 to being annihilation operators , then this equation in just that of the quantum hamiltonian in the heisenberg representation and the time - dependent phase factors are just the time propagators of the creation and annihilation operators .
in particular , for fixed @xmath72 and @xmath195 the off - diagonal coupling terms are of the form , for @xmath23 hermitian and real @xmath196.\ ] ] if one now sets @xmath197 corresponding to classical amplitudes @xmath198 or equivalently heisenberg operators , one obtains coupling elements @xmath199 which are those of eq .
( [ eq : classham ] ) . by contrast , if one restricts the coupling to terms @xmath200 as in the q - coupled case , one has @xmath201 . \end{split}\ ] ] then one sees that to obtain the hamiltonian of eq . ( [ eq : quantform ] ) which maps exactly to the schrodinger equation , it is necessary to neglect the `` non - rotating wave '' terms involving the rapidly oscillating phase factors @xmath202 , which contribute weakly to transition probabilities .
more exactly , let us take @xmath138 to be the value of the largest of the elements @xmath6 i.e. @xmath203 where all @xmath204 are less than unity
. then we can define a dimensionless time as @xmath205 .
the rotating wave factors are then of the order of @xmath206 , where @xmath207 is the mean frequency difference and the non - rotating wave terms are of the order of @xmath208 , where @xmath89 is the mean frequency . now
if @xmath209 and @xmath210 , which is the rca , again we see that classical and quantum couplings will be the same in that the non - rotating wave terms can be neglected .
+ a rather different view of the rca is obtained from the hamilton equations .
the resulting q - coupled newton equations are @xmath211 which are identical in form to eq .
( [ eq : coupclass ] ) for standard classical oscillators .
by contrast , for the p&q - coupling one has @xmath212 which differ from the newton equations by the last term .
however , again scaling as above we have @xmath213 since in rca we have @xmath214 we can neglect the second order term so that the p&q - coupled hamiltonian gives the same second - order classical equations of motion as the q - coupled hamiltonian .
as above , for the moment we ignore friction of resistance and consider an @xmath215 oscillator circuit . for a single oscillator the balance of e.m.f . from inductance and capacitor
is expressed by , @xmath216 where @xmath28 is the charge and @xmath217 is the current related by @xmath218 then we have the harmonic oscillator equation @xmath219 or , @xmath220 where the frequency is @xmath221 . note that , compared to a mechanical oscillator , @xmath222 plays the role of mass , @xmath223 that of coupling constant and @xmath224 corresponds to momentum .
now we couple two identical @xmath215 oscillators by a capacitor @xmath225 connected by leads in which current @xmath226 flows .
the equations of the coupled circuits are , @xmath227 with the e.m.f .
balance @xmath228 using @xmath229 one can eliminate @xmath230 to obtain , @xmath231 where the dimensionless ratio @xmath232 is defined .
these equations can be put also in the form of eqs .
( [ eq : phieqns ] ) , @xmath233 clearly when @xmath234 the equations become those of uncoupled oscillators . adding and subtracting either of these two sets of equations leads to uncoupled equations in the new variables @xmath121 , i.e. @xmath235 which are identical to eqs .
( [ eq : q_pm ] ) except that now @xmath236 and @xmath237 . for coupled @xmath215 oscillators
the rca corresponds to @xmath238 to give , @xmath239 comparison with eq .
( [ eq : omegapprox ] ) shows that in this case we make the identification @xmath240 .
then the coupled electrical oscillator equations giving @xmath241 are identical to the q - coupled pendula equations and , as we have shown , in rca reproduce the complex amplitudes of the quantum dimer .
to include damping phenomenologically in the classical case we add a velocity - dependent term to the oscillator equations ( [ eq : coupclass ] ) , ignoring the corresponding fluctuations , i.e. @xmath242 then the coupled equations ( [ eq : ticlass ] ) become , @xmath243 this equation is written in the form , @xmath244 again , in the rca , we consider that the spread in eigenfrequencies @xmath90 and @xmath43 is small compared to the mean natural frequency @xmath89 to approximate the above equations by , @xmath245 if we assign each quantum level a width @xmath246 then the quantum coupled equations become , @xmath247 which are identical in form to the classical equations . in ref .
@xcite it is shown , how pure dephasing can be realized in a system of coupled oscillators which reproduces the quantum results . | in the first days of quantum mechanics dirac pointed out an analogy between the time - dependent coefficients of an expansion of the schrdinger equation and the classical position and momentum variables solving hamilton s equations .
here it is shown that the analogy can be made an equivalence in that , in principle , systems of classical oscillators can be constructed whose position and momenta variables form time - dependent amplitudes which are _ identical _ to the complex quantum amplitudes of the coupled wavefunction of an @xmath0-level quantum system with real coupling matrix elements .
hence classical motion can reproduce quantum coherence . |
the presently most accurate way to measure the polarized gluon distribution function in the nucleon is to study those processes which can be calculated in the framework of perturbative qcd ( pqcd ) , i.e. for which the involved production cross sections and subprocess asymmetries can be predicted .
one of the cleanest ways to probe qcd is to investigate heavy quarkonia production processes .
heavy quark pair production processes can be controlled perturbatively due to large mass of constituents . on the other hand ,
heavy quark systems are mainly produced in gluon fusion processes and therefore , asymmetries are expected to be sensitive to the polarized gluon distribution in the nucleon .
investigation of heavy quarkonia production processes in polarized experiments would also yield additional information about the quark - antiquark pair hadronization phase .
the two - spin asymmetry in @xmath0 production has been studied in the framework of the so called color singlet model ( csm ) @xcite by morii and collaborators @xcite .
but as was shown in the last years , the color singlet model does not describe satisfactorily the heavy quarkonium hadroproduction at tevatron and also at fixed target energies .
while the csm gives a reasonable description of the @xmath0 production cross section distribution shapes over @xmath4 or @xmath5 at fixed target energies , it completely fails in the explanation of the integrated cross section ( a k factor 7 - 10 is needed to explain experimental data ) @xcite .
the anomalously large cross section @xcite of the @xmath0 production at large transverse momenta at the tevatron revealed another negative feature of the csm . within the framework of the csm it is impossible to explain the anomalously large @xmath6 @xcite and direct @xmath0 production @xcite in the cdf experiment at the tevatron .
the csm is a nonrelativistic model where the relative velocity between the heavy constituents in the bound state is neglected . but
discrepancies between experimental data and the csm predictions hint that @xmath7 corrections as well as other mechanisms of quarkonium production , which do not appear in the leading order in @xmath8 , should be considered .
expansion of quarkonium cross sections and decay widths in powers of relative velocity @xmath8 of heavy quarks in a bound state has recently been realized in terms of nonrelativistic qcd ( nrqcd ) @xcite .
this formalism implies not only color - singlet processes but the new color - octet mechanism , when a quark - antiquark pair is produced on small time scales in color octet states and evolves into a hadron by emission of soft gluons .
the color octet mechanism takes into account the complete structure of the quarkonium fock space while in the csm only the dominant fock state is considered , which consists of a color singlet quark antiquark pair in the definite angular - momentum state ( higher order fock states are suppressed by powers of @xmath8 ) . according to the factorization approach based on the nrqcd , the production cross section for a quarkonium state h in the process @xmath9 can be written as @xmath10\langle0|{\cal o}^{h}[n]|0\rangle}\nonumber\end{aligned}\ ] ] where @xmath11 is the distribution function of the parton @xmath12 in the hadron @xmath13 .
the subprocess cross section is separated into two parts : short distance ( @xmath14 $ ] ) coefficients and long distance matrix elements @xmath15|0\rangle$ ] .
the @xmath14 $ ] is the production cross section of a heavy quark - antiquark pair in the @xmath12 and @xmath16 parton fusion .
it should be calculated in the framework of pqcd .
the [ n ] state can be either a color singlet or a color octet state .
the @xmath15|0\rangle$ ] describes the evolution of a quark - antiquark pair into a hadronic state .
these matrix elements can not be computed perturbatively .
but the relative importance of long distance matrix elements in powers of velocity @xmath8 can be estimated by using the nrqcd velocity scaling rules @xcite .
shapes of the @xmath4 distribution of short distance color - octet matrix elements indicate that the new mechanism can explain the tevatron data of direct @xmath0 and @xmath17 production at large @xmath4 @xcite .
but unlike color - singlet matrix elements connected with the subsequent hadronic nonrelativistic wave functions at the origin , color octet long distance matrix elements are unknown and should be extracted from experimental data . the color octet contribution to the @xmath0 photoproduction has been analyzed in the papers @xcite .
recently , the @xmath0 hadroproduction at fixed target energies has been studied by including the color - octet mechanism @xcite .
large discrepancies between experimental data and the csm predictions for the total cross section of the @xmath0 hadroproduction were explained .
the color octet contribution is dominant in the @xmath0 hadroproduction at energies @xmath18 gev .
the analyses carried out in these papers @xcite demonstrate that fitting the photoproduction and hadroproduction data at low energies requires smaller values than those extracted from prompt @xmath0 production at cdf at the tevatron @xcite .
possible reasons for such discrepancies have recently been analyzed in the papers @xcite .
the extraction of color - octet long distance matrix elements from the quarkonium production properties in polarized experiments would be an additional test of universality . in the present letter we consider the color octet contribution to the double spin asymmetry of @xmath0 hadroproduction .
unlike the previous calculations @xcite we consider @xmath19 color octet and color singlet pair production in @xmath20 subprocesses to obtain asymmetries at nonzero transverse momenta ( @xmath21 gev ) .
such values of @xmath4 can not be caused by internal motion of partons in the nucleon and hence transverse momentum distributions of production cross section and asymmetries are calculable perturbatively .
the double spin asymmetries in parton collisions are presented in section 2 . since heavy quarkonium
is mainly produced in the gluon - gluon fusion subprocesses , the @xmath0 production asymmetry should be sensitive to the polarized gluon distribution function in the proton .
we have calculated the expected asymmetry of @xmath0 production at hera-@xmath2 , one of the future options of hera @xcite ; an experiment utilizing an internal polarized nucleon target in the polarized hera beam with energy @xmath22 gev would yield @xmath23 gev . for comparison
, we also considered the expected asymmetry of @xmath0 production in similar spin physics experiments at much higher energies at the rhic collider @xcite .
let us discuss the two - spin asymmetry @xmath24 for the inclusive @xmath0 production which is defined as @xmath25 where @xmath26 denotes helicity projection sign on the proton momentum direction .
the cross section of @xmath0 production can be written as @xmath27 where @xmath28 denotes the branching ratio of the corresponding @xmath19 state into @xmath0 .
the production of each state of quarkonium is contributed by both color octet and color singlet states , as in the case of direct @xmath0 production @xmath29 ) \langle0|{\cal o}_8^{j/\psi}(^{2s+1}l_j)|0\rangle}\end{aligned}\ ] ] where the sum is over the states @xmath30 , @xmath31 and @xmath32 .
we consider only the dominant sets of color octet states by the nrqcd velocity expansion for the direct @xmath33 and @xmath34 state charmonium production . in a recent paper @xcite
was considered asymmetry of @xmath0 hadroproduction in the color octet model exploiting only the lowest order subprocesses ( @xmath35 ) over the qcd coupling constant .
subprocesses @xmath36 contribute only to the production of a quarkonium state at zero transverse momentum with respect to the beam axis .
transverse momentum distributions of @xmath19 states and , consequently , the @xmath0 meson are not calculable in the @xmath37 region .
to deal with the experimentally observable quantities ( taking into account the kinematic restriction on the angle with respect to the beam axis at hera-@xmath2 and at collider experiments ) , we consider the subprocess @xmath20 which gives the leading contribution to the quarkonium production with @xmath4 greater than @xmath38 gev .
such large transverse momenta can not be caused by internal motion of partons in the nucleon and , respectively , subprocesses @xmath36 should not contribute to the production of quarkonia with @xmath21 gev . for calculating the expected asymmetries we consider only the following subprocesses : @xmath39 for the color octet and singlet states of a @xmath19-pair .
the quark - antiquark collision subprocesses are not taken into account , because of small sea quark polarization in the proton . using the symbolic manipulation program form @xcite
, we have calculated the @xmath19-states production cross sections for different helicity states of colliding partons . for the color
octet states total cross sections have been calculated by cho and leibovich @xcite . the total cross sections for color singlet states are presented in @xcite .
the latter results served for checking our calculations .
( 14,14 ) ( 0,0 ) figure 1 .
partonic level double spin asymmetries for @xmath40 and @xmath41 color octet and @xmath42 color octet and singlet states versus @xmath43 and @xmath44 .
figs.1 and 2 present the values of the subprocess level asymmetries for the color octet and color singlet states production of a ( @xmath45)-pair depending on the two dimentionless quantities : @xmath46 and @xmath47 , where @xmath48 is the maximum momentum of the produced state in the subprocess . in figs.1 and 2
are shown only the gluon fusion subprocess asymmetries because they give the main contribution to the hadronic level asymmetry .
the cases of @xmath40 and @xmath41 singlet states are omitted in figs.1 and 2 .
the @xmath40 state does not contribute to @xmath0 production and analytical expressions for the corresponding cross sections for the @xmath41 state are given in @xcite .
( 14,14 ) ( 0,0 ) figure 2 .
partonic level double spin asymmetries for @xmath49 and @xmath50 color octet and singlet states versus @xmath43 and @xmath44 .
as can be seen from figs.1 and 2 , the asymmetries are very similar for color octet and color singlet states with the same spin - orbital quantum numbers .
it is worth mentioning that in the limit @xmath51 ( i.e. at the threshold of heavy quark pair production ) @xmath52 for the scalar and tensor states tends to 1 and -1 , respectively .
the limit @xmath53 means that the emitted gluon in the @xmath20 subprocess ( @xmath54 ) becomes soft and the helicity properties of the amplitude should be the same as those of the amplitude of the @xmath36 process ( @xmath55 ) .
it is easy to show that the asymmetries for the process @xmath36 for scalar and tensor states are 1 and -1 , respectively .
the existence of such limits serves as an additional test for our analytical calculations of cross sections @xmath56 . for scalar and tensor states
are 1 and -1 , respectively .
the existence of such limits serves as an additional test for our analytical calculations of cross sections @xmath56 .
for the calculation of the hadronic level asymmetries of @xmath0 production we used the long distance matrix elements fitted from various experimental data .
all color singlet matrix elements are related to the radial quarkonium wave functions at the origin and their derivatives . as in paper @xcite for this purpose
we used the buchmller - tye wave functions values at the origin tabulated in ref.@xcite .
the color octet matrix elements were fitted from the @xmath0 and higher charmonium state production data in various experiments .
unfortunately , there are some discrepancies between the values of color octet matrix elements extracted from different experiments
. the number of color octet long distance matrix elements should be reduced by using the nrqcd spin symmetry relations : @xmath57 these relations are accurate up to @xmath58 .
after using these relations we have only four independent matrix elements @xmath59 , @xmath60 , @xmath61 and @xmath62 , which give the main contributions to the @xmath0 hadroproduction cross section . for the first two parameters we used the values extracted from tevatron data by cho and leibovich @xcite .
the other parameters , connected with the @xmath17 production , were also taken from @xcite .
as concerns the parameters @xmath63 and @xmath64 , it is possible to extract only their combination from direct @xmath0 production data at tevatron @xcite : @xmath65 values for the other combinations are extracted from the @xmath0 photoproduction and fixed target hadroproduction data @xcite @xmath66 if one assumes that @xmath67 the photoproduction and fixed target hadroproduction values ( low energy values ) are an order smaller than the tevatron value ( high energy value ) .
the possible sources for such a discrepancy are discussed in @xcite . in paper
@xcite it is mentioned that the mass of the produced hadronic final state ( or intermediate color octet state ) must be higher than that of the @xmath0 meson @xmath68 , because the intermediate octet state emits gluons with energy @xmath69 before transition into a color singlet state . but a charmonium is not a truly nonrelativistic system , the average relative velocity of constituents in @xmath0 is not very small @xmath70 and emitted gluons in the @xmath45 pair hadronization phase have the energy @xmath71 gev .
therefore the mass of the hadronic final state is about @xmath72 gev . at fixed target energies ,
increasing the mass of the produced hadronic state leads to a reduced partonic luminosity because of a steeply falling gluon distribution function .
consequently , the production cross section of the ( @xmath45 ) color octet state should be smaller .
so , the true color octet long distance matrix elements must be larger , than those extracted by using @xmath73 as mass of the intermediate octet state @xcite .
another possible source was mentioned in the paper @xcite and should be an uncertainty connected with the choice of different parametrizations for the gluon distribution function in fitting the direct @xmath0 production data at cdf @xcite .
cho and leibovich used in their calculations the mrsd0 parametrization for parton distribution functions . at small values of @xmath74 gev , using a more reliable parametrization for small partonic @xmath75 ( grv lo or grv ho@xcite ) leads to @xmath76 times higher cross sections for the @xmath42 and @xmath40 color octet state production than those obtained by using mrsd0 parametrization @xcite .
hence the fitted value of combination ( [ high ] ) should be approximately @xmath76 times smaller : @xmath77 at large @xmath78 gev all parametrizations give practically the same magnitude for the color octet states cross sections .
hence if we use the new combination ( [ my ] ) one needs a larger value of the parameter @xmath79 to explain the experimental data for the @xmath0 cross section at @xmath78 gev : @xmath80 practically the same values for these parameters were obtained by beneke and krmer in a recent paper by fitting the cdf data of direct @xmath0 production and using the grv lo ( 1994 ) parametrization @xcite : @xmath81 for calculations of the hadronic level asymmetries we use three different sets for the three long - distance color octet matrix elements @xmath82 as has been mentioned above , values for the other color octet parameters were taken from @xcite .
for the calculations of the expected asymmetries we assume that the leading term in the combination ( [ my ] ) is the parameter @xmath63 ( or @xmath83 ) .
only wiyh such a radical choice the values of high and low energy parameters should be consistent to each other .
3a presents the expected asymmetries for the three sets of color octet parameters in the case when the parameter @xmath64 tends to zero ( @xmath63 is the leading term in the combinations ( [ high]),([low]),([my ] ) ) . ( 16,7.5 ) ( 0,0 ) ( 8,0 ) figure 3 .
the double spin asymmetries for different sets of color - octet long distance parameters .
solid line corresponds to set @xmath84 , dashed line set a , dotted line set b ; ( a ) for the case when @xmath85 , ( b)for @xmath86 .
dot - dashed line in fig.a corresponds to asymmetry calculated by using values of parameters from @xcite .
3b represents the other radical choice , when the second parameter @xmath63 is zero . for the calculations of hadronic level asymmetries we used the grv lo parametrization for unpolarized distribution functions @xcite and a parametrization proposed by gehrmann and stirling for the polarized parton distribution function ( set a ) @xcite . for the mass of the charm quark the value
@xmath87 gev was taken .
as we can see from fig.3 , the magnitude of the asymmetry is more sensitive to the choice of the leading term in the combinations than to the choice of the particular set of the octet matrix elements .
recently the color octet matrix elements @xmath64 and @xmath63 have been extracted separately from the @xmath0 electroproduction data @xcite .
@xmath88 these values confirm our assumption that the leading term in the combinations is the parameter @xmath63 . in fig.3a
, the dotted - dashed curve corresponds to the asymmetry calculated by using the values of the long - distance parameters ( [ eprod ] ) . for the parameter
@xmath89 a value from set c was used .
as one can see from figs.3(a , b ) , the expected asymmetries for all sets of parameters are practically the same . in the following calculations we shall use the set c for the values of the parameters , assuming @xmath90 .
( 16,7.5 ) ( 0,0 ) ( 8,0 ) figure 4 .
the expected asymmetries at @xmath91 gev .
solid line corresponds gehrmann - stirling polarized parton parametrization , set a ; dashed line , set c @xcite . ( a ) expected asymmetries versus transverce momentum , ( b ) versus pseudorapidity of @xmath0 .
dot - dashed curve in the fig.a corresponds to the color - singlet contribution to asymmetry .
fig.4a shows the expected double spin asymmetries at hera-@xmath2 energies as functions of transverse momentum @xmath0 in the c.m.s .
the solid curve corresponds to the gehrman and stirling parametrization for polarized distribution functions , set a ; the dashed curve , to set c @xcite . in fig.4a
we also display the expected statistical errors @xmath92 at hera-@xmath2 which can be estimated from @xcite @xmath93 this relation has been determined by assuming an integrated luminosity of @xmath94 and beam and target polarizations @xmath95 , @xmath96 @xcite .
the error bars are obtained by using integrated cross sections over bins @xmath97 gev ( for the first three points ) and @xmath98 gev ( for the other two ones ) .
the @xmath0 decay branching ratio into the @xmath99 mode is also included .
the magnitude of asymmetries and expected errors allows one to distinguish between different parametrizations of polarized parton distribution functions .
fig.4b shows expected asymmetries depending on the pseudorapidity of @xmath0 at @xmath100 gev .
statistical error bars correspond to the integrated cross section over bins @xmath101 and @xmath102 gev . as in the previous case ,
the errors are small in a wide range of pseudorapidity interval and give a possibility to distinguish between different parametrizations for @xmath103 .
it is worth mentioning that gluon - quark collision subprocesses give about @xmath104 in @xmath105 .
the dot - dashed curve in fig.4a represents the color - singlet contribution to the asymmetry for the parametrization set a ( only the color singlet contribution in @xmath105 was taken into account ) .
it is clear that the color octet contribution dominates in the expected asymmetries of @xmath0 production .
one may ask , where the such a difference in the shapes of asymmetries is coming from , as the subprocess asymmetries are not sensitive to colour structure .
the main reason for this is the large matrix element @xmath106 .
the production of such a state has a large asymmetry @xmath107 ( at @xmath108 , which is not far from the reality for the considered @xmath4 ) , providing the smooth transition to @xmath109 process , discussed above .
as such a state is absent in csm , the large asymmetry itself is already a sign of a presence of colour - octet contribution .
one of main parameters of the model is the mass of the charm quark .
fig.5 shows the asymmetries of @xmath0 production dependening on @xmath110 for two values of @xmath4 for a large set of polarized parton parametrizations ( set a ) .
as displayed in fig.5 , the expected asymmetries are practically insensitive to the quark mass above @xmath111 gev .
( 16,7.5 ) ( 3,0 ) figure 5 . expected double - spin asymmetries at @xmath91 gev versus mass of charm quark . therefore , the double spin asymmetry of @xmath0 production , unlike the cross section , should be free form uncertainties caused by a unknown mass of the intermediate color octet states .
( 16,7.5 ) ( 3,0 ) figure 6 . expected asymmetries and statistical errors at the rhic for two different energies . for comparison
we have also calculated the expected double - spin asymmetries of @xmath0 production at rhic energies .
the results are given in fig.6 for two different values of energy .
the expected statistical errors are calculated with the anticipated the integrated luminosities at the corresponding energies . as we can see from fig.6 , the expected asymmetry decreases with increasing c.m.s .
the statistical errors are calculated by integration over @xmath4 with bins @xmath97 gev of the differential cross sections .
in this paper we investigated the expected double spin asymmetries of heavy quarkonium hadroproduction at hera-@xmath2 . to deal with experimentally observed quantities
, we considered @xmath0 meson production at nonzero transverse momenta , @xmath21 gev .
unlike the calculations of @xcite , where only the lowest order subprocesses were taken into account ( @xmath36 ) , we considered @xmath0 production in the subprocesses @xmath20 because large values of @xmath4 can not be caused by internal motion of partons .
we have calculated the heavy quark pair color octet and color singlet @xmath33 and @xmath34 states production cross sections for different helicities of colliding partons .
the fortran codes for calculated cross sections are available by e - mail . for calculation of the expected hadronic asymmetries we used more reliable values for the color - octet long distance matrix elements , which are in good agreement with those extracted from the @xmath0 electroproduction data @xcite .
the magnitude of expected asymmetries and statistical errors at hera-@xmath2 allows one to distinguish between different parametrizations for polarized parton distribution functions ( gehrmann and stirling , set a and c ) . on the other hand , measuring the asymmetry would give a possibility to extract information about the color - octet long distance matrix elements and to check universality of factorization .
we also calculated the @xmath0 production asymmetries at rhic energies . by comparing the magnitudes of the expected asymmetries at hera-@xmath2 and star ,
it becomes clear that the energy of the fixed target experiment is more preferable for the investigation of the charmonium production asymmetry .
t. morii , s .tanaka , and t. yamanishi , kobe university preprint kobe - fhd-93 - 08 ( 1993 ) , hep - ph/9309336 ( 1993 ) .
schub et al .
phys . rev .
* d52 * ( 1995 ) 1307 .
f. abe et al . , phys .
lett . , 69 ( 1992 ) 3704 and phys .
lett . , 71 ( 1993 ) 2537 .
e. braaten and s. fleming , phys .
lett . * 74 * ( 1995 ) 3327 .
m. cacciari , m. greco , m. mangano and a. petrelli , phys . lett .
* 306b * ( 1995 ) 560 .
bodwin , e. braaten , and g.p .
lepage , phys . rev .
* d51 * ( 1995 ) 1125 .
lepage , l. magnea , c. nakhleh , u. magnea , and k. hornbostel , phys . rev .
* d46 * ( 1992 ) 4052 .
p. cho and a.k .
leibovich , phys.rev . *
d53 * ( 1996 ) 150 , + p. cho and a.k .
leibovich , phys.rev . *
d53 * ( 1996 ) 6203 .
n. cacciari and m. kramer , phys.rev.lett .
* 76 * ( 1996 ) 4128 .
matteo cacciari and michael kramer , phys.rev.lett.76 ( 1996 ) 4128 , + j. amundson , s. fleming and i. maksymyk , uttg-10 - 95 , hep - ph/9601298 .
s. gupta and k. sridhar , tifr / th/96 - 04 , hep - ph/9601349 . m. beneke and i.z .
rothstein , slac - pub-7129 , ucsd-96 - 05 , hep - ph/9603400 .
l. slepchenko and a. tkabladze , hep - ph/9608296 .
nowak , desy 96 - 095 , hep - ph/9605411 .
s. gupta and p. mathews , preprint tifr / th/96 - 53 ( 1996 ) , hep - ph/9609504 g. bunce et al . , particle world * 3 * ( 1992 1 . j.a.m .
vermaseren , symbolic manipulation with form , published by can , kruislaan 413 , 1098 sj amsterdam , 1991 , isbn 90 - 74116 - 01 - 9 .
r. baier and r. rckl , z.phys .
* c19 * ( 1983 ) 251 .
r. gastmans , w. troost , and t.t .
wu , nucl.phys .
* b291 * ( 1987 ) 731 .
m. glck , e. reya , and a. vogt , z.phys . * c67 * ( 1995 ) 433 .
m. beneke and m. krmer , cern preprint cern - th/96 - 310 , ral-96 - 092 , hep - ph/9611218 ( 1996 ) . | we calculated the color - octet contribution to the double spin asymmetry of @xmath0 hadroproduction with nonzero transverse momenta at fixed target energies @xmath1 gev .
it is shown that the color - octet contribution is dominant in the asymmetries .
the expected asymmetries and statistical errors in a future option of hera with longitudinally polarized protons at hera-@xmath2 should allow one to distinguish between different parametrizations for the polarized gluon distribution in the proton .
= 14pt # 1 # 2 # 3 _ int .
j. mod .
phys .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ mod .
phys .
lett .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ nuovo cim .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ nucl .
phys .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ phys .
lett .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ phys .
rev .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ phys . rep .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ phys . rev .
lett .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ rev .
mod .
phys .
_ * # 1 * ( # 2 ) # 3 # 1 # 2 # 3 _ z. phys .
_ * # 1 * ( # 2 ) # 3 jinr e2 - 96 - 431 + hep - ph/9612301 * double - spin asymmetry of @xmath0 production in polarized pp + collisions at hera-@xmath3 * + + _ bogoliubov laboratory of theoretical physics , _ + _ jinr , dubna , moscow region , 141980 , russia _ + |
in the yang - mills theory ghost field lagrangian does nt possess gauge symmetry and together with a gauge fixing term breaks gauge invariance of the effective action .
it makes problems for regularization and renormalization of the theory .
the slavnov - taylor identities are more complicated then the ward identities in electrodynamics . recently a new formulation of the yang - mills theory
was proposed @xcite in which the gauge invariance is broken only by the gauge fixing term . in the paper @xcite it was shown that in the framework of perturbation theory this model is equivalent to the usual yang - mills theory . calculating the observables in this model one can pass to the lorentz - type gauge in which the renormalizability is evident .
nevertheless a proof of renormalizability directly in the gauge proposed in the paper @xcite is absent . in the present paper we calculate at one - loop the gauge field and @xmath2-vertex renormalization constants .
it is shown that a relationship between this constants is the same as in the ordinary yang - mills theory .
let introduce notations .
the lagrangian proposed in @xcite looks as follows @xmath3 the fields @xmath4 , @xmath5 are commuting complex @xmath6-doublets and @xmath7 , @xmath8 are anticommuting ones .
@xmath10 is the covariant derivative ; @xmath11 are pauli matrices .
the yang - mills lagrangian is @xmath12 , @xmath13 .
let introduce the fields @xmath14 , @xmath15 . in terms of these fields
the gauge transformation leaving the lagrangian ( [ l_origin ] ) invariant is @xmath16 being quantized in the lorentz gauge @xmath17 the model is described by the effective action which contains the gauge fixing term and non gauge invariant faddeev - popov ghost lagrangian .
however in this model the lorentz invariant condition @xmath18 is also an admissible gauge . in this case
the gauge invariance is broken by only the gauge fixing term : @xmath19 , here @xmath20 is the lagrangian ( [ l_origin ] ) .
the lagrangian ( [ l_origin ] ) at the surface @xmath18 looks as follows @xmath21 the propagators and vertices are given in the appendix .
we start with considering the gauge field renormalization constant @xmath1 and @xmath2-vertex renormalization constant @xmath0 and show that in the one - loop approximation these constants coincide with the corresponding ones in the ordinary yang - mills theory .
we also consider anticommuting fields renormalization constant @xmath22 and @xmath23-vertex renormalization constant @xmath24 and show that the relationship @xmath25 holds in one - loop approximation .
the calculations can be found in the appendix .
the dimensional regularization is used , hence all tadpole diagrams are equal to zero .
the constant @xmath1 renormalizes the transversal part of one - particle irreducible diagrams with two external gauge field lines @xmath26 .
there are three types of the contributions to the divergent part of the corresponding diagrams .
the first one is a contribution of gauge field @xmath26 loops .
this contribution is the same as in the ordinary yang - mills theory , because the vertices @xmath2 , @xmath27 and the propagators coincide with the usual ones .
the second one is the contribution of loops with commuting scalar fields @xmath28 , @xmath29 .
and the third one is the contribution of loops with anticommuting fields @xmath7 , @xmath8 .
the sum of the second and the third contributions equals to @xmath30 in the ordinary yang - mills theory divergent part of faddeev - popov ghosts loop is the following @xcite , @xcite .
@xmath31 one can see that transversal parts of ( [ z2_sumcommanticomm ] ) and ( [ z2_fpghosts ] ) coincide . as it was mentioned above the contributions of gauge field @xmath26 loops in the present theory and in the yang - mills theory
coincide and therefore constants @xmath1 coincide too . note that contrary to the standard formulation of the yang - mills theory the radiative corrections generate also renormalization of the longitudinal part of the polarization operator , that is the counterterm proportional to @xmath32 . such a counterterm does not break the gauge invariance , because it originates from the admissible gauge invariant counterterm @xmath33 after the shift @xmath34 .
the constant @xmath0 performs renormalization of @xmath2-vertex .
again there are three types of contributions .
contribution of the gauge field @xmath26 loops coincide with the corresponding one in the ordinary yang - mills theory . in one - loop approximation
the total contribution of commuting scalar fields @xmath28 , @xmath29 and anticommuting fields @xmath7 , @xmath8 is equal to @xmath35 it coincides with the contribution of faddeev - popov ghosts in the ordinary yang - mills theory and therefore the constant @xmath0 equals yang - mills constant @xmath0 .
thus in one - loop approximation the constants @xmath1 , @xmath0 are the same as in the yang - mills theory and therefore the expression @xmath36 is the same as in the yang - mills theory . @xmath37 which means that the model is asymptotically free . to find the constants @xmath22 , @xmath24 ,
it is necessary to calculate the divergent part of the diagrams represented on fig .
[ fig_z2bar ] , [ fig_z1bar ] . performing the calculations we obtain @xmath38 thus @xmath39 .
in the present paper it was shown that the theory with the gauge invariant ghost lagrangian agrees with the standard yang - mills theory in one - loop approximation . the relationship @xmath40 between renormalization constants which is one of the necessary conditions of gauge invariance of renormalized lagrangian holds
. a complete renormalization of the theory requires further investigation .
* acknowledgements * + i am grateful to a.a .
slavnov for useful discussion and remarks .
99 a.a.slavnov , a lorentz invariant formulation of the yang - mills theory with gauge invariant ghost field lagrangian , jhep 08(2008)047 l.d.faddeev,a.a.slavnov gauge fields .
introduction to quantum theory .
second edition .
addison - wesley publishing company 1991 .
chapter 4 .
m.e.peskin , d.v.schroeder an introduction to quantum field theory .
addison - wesley publishing company 1995 .
chapter 16 .
here we present the calculations of the commuting scalar fields @xmath28 , @xmath4 and anticommuting fields @xmath7 , @xmath8 loops contributions to the divergent part of the gauge field two - point green function . for calculation of commuting scalar fields contribution
it is necessary to find divergent part of two diagrams which are represented at the fig .
[ fig_z2commute ] . left diagram ( fig .
[ fig_z2commute ] ) gives the following integral .
@xmath41 it is easy to see that the right diagram at fig .
[ fig_z2commute ] coincides with the left diagram after the change @xmath42 , @xmath43 , @xmath44 and consequently it gives the same result .
therefore the total commuting scalar fields contribution is @xmath45 now let us find the anticommuting fields contribution .
we calculate divergent part of diagrams shown at fig .
[ fig_z2anticommute ] . for the first diagram ( fig .
[ fig_z2anticommute ] ) we have @xmath46 one can see that the remaining diagrams have identical divergent parts .
furthermore sum of these divergences is equal to the divergence of the first diagram .
therefore the total anticommuting fields contribution is @xmath47 summarizing ( [ app_z2_commute ] ) and ( [ app_z2_anticommute ] ) we find @xmath48
now we calculate the contribution of the commuting scalar fields @xmath28 , @xmath4 and the anticommuting fields @xmath7 , @xmath8 into divergent part of the gauge field three - point green function .
the contribution of commuting fields @xmath28 and @xmath29 loops is given by the sum of six diagrams one of which is represented at fig .
[ fig_z1commute ] and other five diagrams differ from this one by permutation of vertices .
the diagram at fig .
[ fig_z1commute ] gives @xmath49 contributions of five remaining diagrams can be obtained from formula ( [ z1_commute1of6 ] ) .
for example if we replace @xmath50 , @xmath51 , @xmath52 , @xmath53 in this formula , we obtain the divergent part of the diagram which differs by permutation of two bottom vertices . summarizing all six we obtain @xmath54
now let us consider the loops of the anticommuting fields @xmath7 and @xmath8 .
corresponding diagrams are presented at fig .
[ fig_z1anticommute ] . vertices @xmath55 ,
@xmath56 , @xmath57 differ from each other only by tensors @xmath58 , @xmath59 , @xmath60 correspondingly . for the top and bottom diagrams we obtain correspondingly @xmath61
so these diagrams give the same results .
in addition one can see that each remaining diagrams give the same result too .
so we have to calculate divergent part of the first diagram at fig . [ fig_z1anticommute ] and multiply it by eight
. @xmath62 the total contribution of the anticommuting fields loops and loops of the fields @xmath28 , @xmath29 is @xmath63
now we calculate the constants @xmath64 , @xmath65 .
let us consider the one - loop diagram at fig .
[ fig_z2bar ] .
the corresponding integral is equal to @xmath66 this divergence is canceled by the following counterterm : @xmath67 to find the constant @xmath24 one has to calculate the sum of divergent parts of diagrams which are presented at fig .
[ fig_z1bar ] .
besides the presented diagrams there are the diagrams with three vertices , but their divergent parts are equal to zero . for the left diagram at fig .
[ fig_z1bar ] we obtain the following integral .
@xmath68 for the right diagram on fig . [ fig_z1bar ]
we obtain @xmath69 summarizing the results ( [ z1bar_left ] ) and ( [ z1bar_right ] ) one gets @xmath70 this divergencies are canceled by counterterm @xmath71 . | one - loop calculations of renormalization constants @xmath0 , @xmath1 in the model , proposed in the paper @xcite with gauge invariant ghost field lagrangian are performed .
it is shown that the model is asymptotically free and the renormalization constants satisfy the same equation as in the ordinary yang - mills theory . |
a central feature of shannon s @xcite theory of information is the notion of probability distribution , that is , a convex linear combination of outcomes of some experiment . replacing probability distributions with quantum superpositions , that is , complex projective linear combinations of outcomes of the experiment , and then following by measurement has important consequences for the treatment of information in the quantum realm .
the resultant generalization of shannon theory of information underpins many of the important applications of quantum mechanics , such as quantum computing , and the much broader theory of quantum information processing that embeds the classical treatments of probability distributions applied to observables within that of positive - definite , trace preserving maps .
much of the work on quantum games to date has focused on the issue of what quantum mechanics brings to game theory .
the perspective is decidedly that quantum mechanics can offer some potential advantage , such as being able to resolve game - theoretic conundrums ( one such being that of the prisoner s dilemma ) , by the utilization of quantum resources unavailable within a classical framework of the game .
whilst this focus is interesting , it is our belief that game - theory is at its most fruitful when applied as an analysis tool for situations in which there may be an underlying competitive element .
game theory has found such fruitful applications in fields as diverse as economics and biology .
here we are interested in whether the tools of game theory will allow us to understand , or design , quantum physical processes in a different way . in our previous work in @xcite
we have characterized this approach as that of `` gaming the quantum '' rather than quantizing a game @xcite .
quantum computation has been shown to behave differently than classical computation .
well known examples of this different behavior of quantum computation include grover s quantum search algorithm @xcite for identifying an element from a list which outperforms its classical counterpart by a factor that is quadratic in the size of the list , and shor s @xcite factoring algorithm that offers an exponential speed up over classical algorithms for factoring large integers .
meyer @xcite has posed the question whether game theory can be used to gain a different insight into quantum computation and , in particular , whether the notion of nash equilibrium applied to quantum computation might be a useful tool in the design of quantum algorithms . with this in mind , a game - theoretic treatment of two qubit quantum computations as two player , strictly competitive games is presented here in section [ sec : quantumgames ] . in particular , we identify nash equilibrium outcomes in these quantum games as characterized by the geometry of the hilbert space of quantum computations , an idea we first presented in @xcite .
expanding on this earlier work , we show here that this geometric characterization implies the equivalence of nash equilibrium and mini - max outcomes in quantum games of this type .
further , in section [ sec : mechanism ] , we design quantum mechanisms that identifies mini - max outcomes in two qubit quantum computations via linear programming .
this approach allows us to consider the output of some quantum process as optimal in the sense of nash equilibrium . in the case of two qubit quantum computations considered here ,
a computation of this type is said to be _ optimal _ if the output it produces is the best possible under the constraints given . the conditions for reverse - engineering such a quantum computation are laid out in generality by the analysis of the mini - max conditions we derive .
we give the example of the implementation of a cnot gate in section [ sec : cnot ] to illustrate this principle .
this perspective also has ramifications when considering the evolution of quantum states in general .
if we consider the maximally entangled state as the output of some quantum game in which the players generate the entangled state as an equilibrium output then we can understand maximal entanglement as a consequence of the `` push - pull '' between the players competing strategies .
our reverse - engineering approach allows us to construct the family of games in section [ sec : entanglementgame ] for which a maximally entangled state is the mini - max output state produced and we illustrate this with an example .
we end this section with an open question : can grover s quantum search type algorithms be cast as strictly competitive quantum games so that the resulting game model offers insights into quantum informational processes ?
a finite non - cooperative quantum game @xmath0 is a function with a @xmath1-dimensional hilbert space @xmath2 of quantum superpositions as its range , together with the additional feature that `` players '' entertain non - identical preferences over the elements of the range . in symbols
@xmath3 the factor @xmath4 in the domain of @xmath0 is the set of _ quantum strategies _ of player @xmath5 and a _ play _ of the game @xmath0 is a tuple of strategies in @xmath6 producing a _ payoff _ to each player in the form of an _ outcome _ , that is , an element of the range of @xmath0 .
if upon some appropriate restrictions @xmath0 reduces to some non - quantum `` classical '' game @xmath7 , then @xmath0 is said to be a _ quantization _ of @xmath7 .
as bleiler points out in @xcite , such quantized games can be viewed as extensions of some classical game @xmath7 into the quantum realm in the same way a _ mixed _
game is an extension of a game into the realm of probabilities . for a detailed treatment of game theory , readers are referred to binmore s excellent book on the subject @xcite . from now on ,
the terms quantum strategy and strategy will be used interchangeably .
nash equilibrium _ is a play of a ( quantum ) game in which every player employs a strategy that is a best reply to the strategic choice of every other players . in other words , unilateral deviation from a nash equilibrium by a player in the form of a different choice of strategy
will produce an outcome that is less than or equal to in preference to _ that _ player than before . in @xcite
, we characterize the notion of nash equilibrium via the inner - product of the hilbert space of quantum superpositions in a two player , _ strictly competitive _ quantum game , that is , a game in which the preferences of the players are diametrically opposite . because of this diametrically opposite nature of the players preferences , nash equilibria in two player , strictly competitive games are plays in which each player minimizes his opponent s maximum possible payoff thus maximizing his own minimum possible payoff .
this feature of nash equilibria in strictly competitive games is known as the mini - max theorem .
more precisely , if @xmath0 is a two player strictly competitive quantum game , then @xmath8 and the players preferences may be defined as follows .
let @xmath9 be an orthogonal basis of @xmath10 corresponding to some observable .
define the preferences of player i over the elements of @xmath11 to be @xmath12 and player ii s preferences to be @xmath13 where the symbol @xmath14 stands for `` strictly preferred over '' and the symbols @xmath15 stands for `` indifferent between '' .
note the diametrically opposite nature of the players preferences with respect to the elements @xmath16 and @xmath17 .
these preferences of the players over the elements of the basis @xmath11 induce preferences over arbitrary quantum superpositions in @xmath10 via the notion of distance or angle between quantum superpositions . to be more precise ,
let @xmath18 and @xmath19 be quantum superpositions in @xmath10 , and let @xmath20 denote the distance between the two quantum superpositions as measured by the angle between them .
then for player i @xmath21 similarly , if @xmath22 and @xmath23 are two quantum superpositions in @xmath10 , then for player ii @xmath24 a play @xmath25 of the quantum game @xmath0 is a nash equilibrium if unilateral deviation by any one player from @xmath25 will produce a quantum superposition that is either equal to in distance or further away from that player s most preferred element of @xmath11 .
that is , if player i unilaterally deviates from the @xmath25 and instead engages in any other play @xmath26 then @xmath27 also , if player ii unilaterally deviates from @xmath28 and instead engages in any other play @xmath29 then @xmath30 inequalities ( [ eqn:1 ] ) and ( [ eqn:2 ] ) characterize a nash equilibrium outcome in a two player , strictly competitive quantum game as a simultaneous minimization problem .
note however that this simultaneous minimization problem is equivalent to the one in which each player minimizes the maximum possible payoff , in terms of the quantity @xmath31 , of the other player .
this is further equivalent to the problem in which each player maximizes his own minimum payoff in terms of the quantity @xmath31
. we therefore have the following classic result from two player , strictly competitive game theory for quantum games of the same type . + * proposition * : _ an outcome of a two player , strictly competitive quantum game @xmath0 is a nash equilibrium if and only if it is a mini - max ( or maxi - min ) outcome . _ + in the broader context of quantum information , the strictly competitive , quantum game @xmath0 described above can be viewed as an information channel processing quantum information with outcomes of this channel marked with respect to non - identical preferences of some background players .
whether this game - theoretic perspective and the notion of nash equilibrium leads to any new insights or results in the area of quantum information processing is an open question ; indeed , it is an undiscovered country . leaving the exploration of this undiscovered country to a subsequent expedition
, we restrict here only to the much simpler case of quantum computation with two qubits .
we point out that _ qubit _ is simply another term for quantum superpositions defined for two possible outcomes of some experiment . as dictated by the axioms of quantum mechanics , the strictly competitive quantum game is now necessarily a unitary operator @xmath32 where @xmath33 is the hilbert space of a qubit designated here as each player s set of quantum strategies , and @xmath34 , the hilbert space of the _ joint states _ of two qubits , is designated as the set of outcomes . at this point
, the theory of hilbert space guarantees that the image of the unitary operator @xmath35 is a sub - hilbert space of @xmath34 .
moreover , for a given sub - hilbert space @xmath36 of a hilbert space @xmath37 , there exists a unique element @xmath38 that minimizes the `` distance '' between elements of @xmath36 and a fixed @xmath39 that can be measured by @xmath40 .
therefore , with @xmath41 , and @xmath42 , it is guaranteed that there exist @xmath43 that minimize the distance between @xmath44 , respectively .
we seek here a _
mini - maximizer _ , that is , @xmath45 such that @xmath46 . in the following section ,
we take a mechanism design approach to construct a two player , strictly competitive quantum game @xmath35 and the strategic choices of the players in @xmath35 that realize mini - maximizers .
in this section , we take the quantum game @xmath35 to be a generic @xmath47 unitary matrix or , in other words , a generic two qubit quantum logic gate , and explore features of this game and the quantum strategic choices that the players will make in this game in order to observe a mini - max outcome .
let @xmath48 be the standard ( orthogonal ) computational basis for @xmath34 , with @xmath49 and preferences of the players defined as in expressions ( [ eqn : pref i ] ) and ( [ eqn : pref ii ] ) and let @xmath50 be a strictly competitive , two player , two strategy quantum game represented with respect to the basis @xmath11 .
also , let @xmath51 be the computational basis for each @xmath52 that constitutes the players set of strategies .
a strategy for player i is the choice of a qubit state , say @xmath53 and a strategy for player ii is the choice of qubit state , say @xmath54 note that these quantum strategic choices can be physically implemented via local unitary operations @xmath55 and @xmath56 on the qubit state @xmath57 respectively .
also note that the notion of quantum strategy depends on the `` choice '' of some initial state on which players act with appropriate unitary operators .
this feature of quantum strategies , as defined here , of acting on some initial state arises from the underlying physics of the quantum system being gamed and is an idea not typically entertained in the toy game models used in game theory .
next , consider a mini - max play of the quantum game @xmath35 , that is , a play in which player i chooses the strategy @xmath58 and player ii chooses the strategy @xmath59 the output of the game is then @xmath60 with @xmath61 and @xmath62 @xmath63 a unilateral deviation on part of player i from the play @xmath64 to _ any _ other @xmath65 will produce the output state @xmath66 such that @xmath67 with @xmath68 because the inverse cosine function is decreasing , it follows that @xmath69 furthermore , because the quadratic function is one - to - one and increasing for non - negative inputs , it follows that @xmath70 likewise , a unilateral deviation on part of player ii from the play @xmath64 to _ any _ other @xmath71 will produce the output state @xmath72 such that @xmath73 such that @xmath74 factoring and applying triangle inequality on the right - hand side of inequality ( [ eqn : ineq1 ] ) reduces it to @xmath75 which can be expressed compactly as @xmath76 with @xmath77 and @xmath78 . applying a similar reasoning to inequality ( [ eqn : ineq2 ] )
produces @xmath79 with @xmath80 and @xmath81 .
inequalities ( [ eqn : shortineq ] ) and ( [ eqn : shortineq1 ] ) can be further simplified by factoring and applying the triangle inequality to their respective left - hand sides , giving two cases each : @xmath82 or @xmath83 and @xmath84 or @xmath85 the quantum game @xmath35 will entertain a mini - max outcome when inequalities ( [ eqn : case1 ] ) and ( [ eqn : case3 ] ) are satisfied , or ( [ eqn : case1 ] ) and ( [ eqn : case4 ] ) , or ( [ eqn : case2 ] ) and ( [ eqn : case3 ] ) , or ( [ eqn : case2 ] ) and ( [ eqn : case4 ] ) are satisfied .
to analyze the first case of mini - maximizer as determined by inequalities ( [ eqn : case1 ] ) and ( [ eqn : case3 ] ) , first rewrite inequality ( [ eqn : case1 ] ) as @xmath86 when @xmath87 .
geometrically , the family of solutions of inequality ( [ eqn : rewrite ] ) consists of all those sets of points in @xmath88 such that the points are on or above the line @xmath89 , with @xmath90 as the independent variable and @xmath91 as the dependent variable , and which is also confined to lie within the unit circle @xmath92 . the condition @xmath87 necessary to derive inequality ( [ eqn : rewrite ] ) can be made game - theoretically meaningful and quantum mechanically insightful by viewing it as an artifact of the quantum game @xmath35 that manifests via appropriate conditions on the quantities @xmath93 and @xmath94 , in contrast to the view that it is an artifact of special assumptions regarding the parameters @xmath95 or @xmath96 that define player ii s choice of strategy at mini - max .
this view keeps the mini - max analysis as general as possible .
a similar analysis applies to player ii s strategic choice at mini - max via the inequality @xmath97 derived from inequality ( [ eqn : case3 ] ) when @xmath98 .
as before , geometrically , the family of solutions of inequality ( [ eqn : rewrite1 ] ) consists of all those sets of points in @xmath88 such that the points are on or above the line @xmath99 , with @xmath96 as the independent variable and @xmath95 as the dependent variable , and which is also confined to lie within the unit circle @xmath100 .
as was done above , the condition @xmath101 necessary to derive inequality ( [ eqn : rewrite1 ] ) can be made game - theoretically meaningful and quantum mechanically insightful by viewing it as an artifact of the quantum game @xmath35 that manifests via appropriate conditions on the quantities @xmath93 and @xmath94 , in contrast to the view that it is an artifact of special assumptions regarding the parameters @xmath91 or @xmath90 that define player i s choice of strategy at mini - max .
again , this view keeps the mini - max analysis as general as possible .
figure [ fig : ne1 ] captures in a generic sense the region that defines nash equilibrium plays in two player , strictly competitive quantum games @xmath35 that satisfy the conditions @xmath87 and @xmath101 .
the horizontal axis represents both @xmath90 and @xmath96 and the vertical axis represents both @xmath91 and @xmath95 .
a play @xmath64 of a strictly competitive two player , two strategy quantum game is a nash equilibrium if the parameters @xmath102 and @xmath103 defining player i s strategy @xmath104 and the parameters @xmath105 and @xmath106 defining player ii s strategy @xmath107 satisfy the constraints of figure 1 .
note that one could instead isolate the quantities @xmath90 and @xmath96 by assuming that @xmath108 and @xmath109 in the quantum game @xmath35 , or indeed isolate @xmath91 and @xmath96 when @xmath87 and @xmath109 and isolate @xmath90 and @xmath95 when @xmath101 and @xmath108 .
any one of these cases will produce a region in @xmath88 bounded by lines that will correspond to mini - max outcomes in those two player , strictly competitive quantum games @xmath35 that satisfy the conditions of the case .
the remaining inequality pairs corresponding to mini - max outcomes can be analyzed similarly .
consider the controlled - not ( cnot ) quantum logic gate @xmath110 as the two player , strictly competitive quantum game @xmath35 discussed above . in this case ,
@xmath111 , @xmath112 , @xmath113 .
inequality ( [ eqn : rewrite ] ) gives @xmath114 however , we can not utilize inequality ( [ eqn : rewrite1 ] ) because @xmath115 .
instead , we choose to rewrite inequality ( [ eqn : case3 ] ) as @xmath116 assuming that @xmath117 .
this gives @xmath118 meaning that a mini - maximizer sees player i employ a quantum strategy for which the length of the @xmath119-coordinate is the maximum possible and player ii employ a quantum strategy for which the length of the @xmath120-coordinate is maximum possible .
this can be achieved by player i choosing the component @xmath121 of his quantum strategy to be any complex number of length 1 and player ii choosing the component @xmath122 of his quantum strategy to be any complex number of length 1 . in other words , for a mini - maximizer in the cnot game based on the conditions of inequalities ( [ eqn : rewrite ] ) and ( [ eqn : rewrite3 ] ) @xmath123 that is , player i chooses a quantum strategy from the subspace of @xmath34 generated by the state @xmath124 whereas player ii chooses a quantum strategy from the subspace spanned by @xmath125 .
reverting back to quantum logic gate perspective , the preceding discussion tells us that the cnot gate performs an optimal quantum computation , in the sense of mini - max , when its input states satisfy the conditions in ( [ cnotresult ] ) .
that is , the cnot gate performs optimally exactly when the input states are separable , a result that is perhaps surprising given that entanglement between states is often expected to be the source of interesting quantum mechanical behavior .
mini - maximizers in the cnot game based on the conditions specified by inequalities ( [ eqn : case1 ] ) and ( [ eqn : case4 ] ) , ( [ eqn : case2 ] ) and ( [ eqn : case3 ] ) , and ( [ eqn : case2 ] ) and ( [ eqn : case4 ] ) can be analyzed similarly .
consider the maximally entangled state @xmath126 as a mini - maximizer @xmath127 in a two player , strictly competitive game in which the corresponding strategic choice of each player is the qubit state @xmath57 , and preferences of the players are identified by ( [ eqn : pref i ] ) and ( [ eqn : pref ii ] ) .
we design quantum mechanisms that will entertain @xmath128 as a mini - maximizer .
note that for @xmath128 , the output state ( [ eqn : output ] ) of the quantum mechanism @xmath35 requires that @xmath129 , from which it follows that @xmath115 .
in this situation , we can not utilize inequality ( [ eqn : rewrite1 ] ) which requires the contrary .
however , note that for @xmath128 , @xmath130 need not be equal to @xmath131 and therefore by assuming that @xmath109 , we can use inequality ( [ eqn : rewrite3 ] ) here to get @xmath132 this last result gives no insight into properties of the quantum mechanism @xmath35 .
however , inequality ( [ eqn : rewrite ] ) gives @xmath133 therefore , the parameter @xmath134 of the quantum game @xmath35 must satisfy inequality ( [ eqn : rewrite4 ] ) in terms of a strategic choice of player i _ other _ than the mini - maximizer .
switching to the quantum computational perspective , the preceding discussion tells us that a quantum mechanism @xmath35 will produce the maximally entangled state ( [ maxent ] ) as a mini - maximizer , with ground state @xmath135 as the input , exactly when its component @xmath134 satisfies inequality ( [ eqn : rewrite4 ] ) for _ any _ value of one of the input qubits other than the ground state .
other quantum mechanisms can be constructed that take the strategic choices of the players corresponding to the ground state to the maximally entangled state considered here as a mini - maximizer , based on the other possible combinations of the inequalities introduced in section [ sec : minimax ] as well as for other possible strictly competitive preferences of the players .
we set up a game - theoretic model for viewing two qubit quantum computations as strictly competitive quantum games and utilized a geometric characterization of the notion of nash -equilibrium in such quantum games developed in @xcite to classify nash equilibria in the form of mini - max outcomes in one instance of these games via regions in @xmath136 , and set up conditions for the classification of mini - max outcomes in other instances .
we propose that a mini - max outcome can be used as a measure of optimality of a quantum computation under strictly competitive constraints such as the one s discussed in section [ sec : quantumgames ] .
gaming the quantum is a natural course of action when exploring the quantum landscape for constrained optimization .
we gamed only the relatively straightforward case of two qubit quantum computations here , modeled as strictly competitive games with a specific choice of preferences for the players ; an extension to this is to investigate these games with different choices of preferences for the players by relaxing the strictly competitive condition . our initial results in this direction suggest that this is richer and more subtle than the strictly competitive case considered here . further , gaming of @xmath137 qubit and continuous variable quantum computations , initially as strictly competitive games and later as more general non - cooperative games , and exploring the nature of nash equilibria more general than mini - max in these computations is another important extension to the work we have presented here .
even further , more general functions permitted in the quantum realm such as those that describe mixed quantum states , might be amenable to gaming . in this context , the question of capacity of a quantum informational channel , a notion that has been described in the past as a mini - max problem in a non - game theoretic context @xcite , might be re - cast as a strictly competitive game .
the emerging area of a quantum neural networks has already been cast as a gaming problem with the notion of mini - max outcomes corresponding to that of optimal learning strategy for the network @xcite .
another important open question of a fundamental nature is whether the evolution of schrdinger s equation might be gamed to obtain fresh insight into how one might speak meaningfully of optimality" of various features of quantum mechanics . | two qubit quantum computations are viewed as two player , strictly competitive games and a game - theoretic measure of optimality of these computations is developed . to this end
, the geometry of hilbert space of quantum computations is used to establish the equivalence of game - theoretic solution concepts of nash equilibrium and mini - max outcomes in games of this type , and quantum mechanisms are designed for realizing these mini - max outcomes . |
thermoelectric effects have generated an immense interest for quite some time already because they offer the possibility to convert heat from the environment into electrical work @xcite .
this form of energy harvesting is potentially useful for electric circuits on modern computer chips that produce large amounts of heat and currently need to be cooled actively in order not to overheat .
one can also use the harvested energy to run auxiliary circuits such as autonomous sensors or recycle the lost energy to charge a battery .
unfortunately , even after decades of material research current thermoelectric materials still have a very low efficiency in converting heat into electrical work and deliver only moderate powers .
for this reason , thermoelectric energy harvesting so far is restricted to certain niche applications such as in interplanetary spaceships where the fact that a thermoelectric generator does not require any moving parts and therefore no maintenance turns out to be useful .
we give a brief summary of how conventional thermoelectric devices work to orient the reader .
the building blocks of thermoelectricity are the peltier and seebeck effect .
the seebeck effect is the flow of electrical current in response to an applied temperature difference , while the peltier effect is the reverse : the creation of temperature difference in response to an applied electrical voltage .
this is intuitively understood by the fact that the hotter charge carriers diffuse faster than the cold ones , creating a flow of thermal energy from hot to cold , and consequently the imbalanced charge builds up an electrical voltage across the material .
one challenge in making good thermoelectric devices is that materials that are good electrical conductors tend also to be good thermal conductors .
this fact makes it difficult to maintain the necessary temperature difference needed to produce electrical power from the seebeck effect .
the need for ways to create systems with high electrical conductance , while maintaining low thermal conductance is an outstanding challenge in this field that the interface - based devices we shall describe can help to solve .
commercial devices use doped semiconductors such as bismuth telluride . in an n - type semiconductor ,
heat flow is in the same direction of the electron ( current ) flow , however , in a p - type semiconductor , heat and current flows are in opposite directions .
this is useful because a thermoelectric architecture can be built of alternating p - type and n - type semiconductor elements that are connected electrically in series ( via metallic contacts ) and thermally in parallel .
this permits the small electrical voltage ( or power ) produced by a single element to be increased by the number of elements in the device , making a practically useful system .
these are placed between ceramics plates which conduct heat , but not electricity , so one side can be heated and the other cooled to produce the power .
alternatively , power can be applied to cool the cold side , and heat the hot one , acting as a refrigerator .
mesoscopic solid - state physics can help to overcome the limitations of current thermoelectric materials by providing powerful and highly efficient heat engines operating at the nanoscale .
soon after initial studies on the thermopower of basic mesoscopic structures like quantum point contacts @xcite and quantum dots @xcite , there were first proposals pointing out that structures of reduced dimension can give rise to an increased thermoelectric figure of merit @xmath0 as compared to bulk structures made from the same materials @xcite .
similar ideas where brought forward by mahan and sofo @xcite who showed that sharp spectral features give rise to high thermoelectric performance as characterized by a high value of @xmath0 .
nanoscale conductors such as quantum dots naturally provide these sharp spectral features .
hence , they are promising candidates for thermoelectric energy harvesters
. initial experiments on quantum dots defined in a two - dimensional electron gas showed saw - toothlike oscillations of the thermopower as a function of gate voltage in agreement with theory @xcite .
further studies have been carried out on open quantum dots @xcite and carbon nanotubes @xcite .
more recently , the thermopower of quantum dots defined in nanowires has been studied @xcite .
interestingly , these experiments could observe nonlinear thermoelectric effects @xcite . in another series of experiments the thermopower due to sequential tunneling , cotunneling and
the kondo effect has been investigated in gate - defined quantum dots @xcite .
the thermopower of double dots was observed in ref .
@xcite .
two - terminal geometries using mesoscopic conductors have been considered , notably using quantum dots @xcite . in the two - terminal geometry ,
both temperature and voltage bias are applied to the sample and the thermoelectric response is investigated .
it has the advantage of being the analogue of the traditional thermocouple which has wide applications , mainly in its role as a thermometer .
however , for the purpose of energy harvesting , it suffers from the fact that different parts of the same electrical circuit must be at different temperatures which makes thermal isolation difficult .
sketch of a generic thermoelectric energy harvester : a mesoscopic system situated at the interface between a cold and a hot bath converts a heat current @xmath1 into a charge current @xmath2 that is able to power a load or charge a battery . ]
in contrast to the above mentioned works based on two - terminal configurations , the problem of a device that powers a circuit by harvesting energy from the outside environment demands a three - terminal geometry , cf .
figure [ fig : harvest ] .
two terminals at the same temperature define the conductor that supports a charge current @xmath2 .
the third terminal represents the coupling to the heat source from which energy but not charge is absorbed .
the harvester converts the heat current @xmath1 into useful power , @xmath3 , that runs a load resistance , represented here by a voltage drop @xmath4 opposed to the generated current .
the properties of the mesoscopic region to which the three terminals are connected define the characteristics of the heat engine , in particular its efficiency , @xmath5 .
only recently has the investigation of thermoelectric effects in quantum - dot structures in three - terminal geometries generated a lot of attention @xcite .
this was motivated by a number of reasons .
first of all , such three - terminal setups share a number of features with mesoscopic coulomb - drag setups that have been studied both theoretically @xcite as well as experimentally @xcite in the last years .
such coulomb - drag setups consist of two nearby mesoscopic conductors such as quantum dots .
the first conductor is subject to a bias voltage that drives a charge current through it .
nonequilibrium charge fluctuations ( noise ) due to this current will induce charge fluctuations in the second , unbiased conductor . if the second conductor is intrinsically nonlinear it will rectify these induced charge fluctuations and thus exhibit a charge current without an applied bias voltage
. similarly , in a three - terminal heat engine , thermal fluctuations from the hot source can get rectified and drive a directed charge current . here
, the nonequilibrium situation is introduced by the temperature imbalance between the hot source and the conductor .
in addition , three - terminal heat engines offer the advantage of spatially separating the hot and cold reservoirs , cf . .
this becomes most pronounced in a recently proposed microwave - cavity heat engine where two mesoscopic conductors such as double quantum dots are connected via a superconducting microwave cavity over a typical length of a few centimeters @xcite .
the separation of hot and cold baths helps to reduce leakage heat currents that are detrimental to achieving a high efficiency of heat to work conversion .
finally , three - terminal heat engines also exhibit a crossed flow of heat and charge currents .
this is useful for energy harvesting applications because it allows the hot source to be kept electrically separated from the actual energy harvester .
thermoelectric configurations including thermometer terminals have also been investigated @xcite .
in contrast to bulk thermoelectric materials , nanoscale heat engines often operate in the nonlinear regime .
linear response theory is valid if the temperature difference on the scale of the inelastic scattering length is small compared to the average temperature @xcite . for bulk materials , the inelastic scattering length is typically much smaller than the system size , such that even for a sizable temperature bias , linear response theory is a good approximation . for nanoscale setups , system size and
inelastic scattering length become comparable .
hence , nonlinear effects become important @xcite . in order to properly describe nonlinear thermoelectric effects in mesoscopic structures , a nonlinear scattering matrix theory of thermoelectric transport
has been developed recently @xcite .
it has been applied to study heat transport through quantum dots where , e.g. , a deviation from the wiedeman - franz law was found @xcite .
it was also used to investigate the magnetic - field asymmetry of nonlinear transport coefficients @xcite .
experimentally , the nonlocal thermoelectric response of a ballistic four - terminal structure has been investigated @xcite .
further studies theoretically analyzed nonlinear thermoelectric transport through molecular junctions @xcite . an important feature of nonlinear thermoelectrics is that the figure of merit @xmath0 is no longer sufficient to characterize the thermoelectric performance @xcite .
instead , one has to rely on quantities such as the maximal efficiency , the efficiency at maximum power @xcite or the maximal efficiency at a given output power @xcite . here
, we review recent theoretical work on three - terminal thermoelectrics with quantum dots .
the review is organized as follows . in sec .
[ sec : coulomb ] we discuss systems of coulomb - coupled conductors ranging from quantum dots in the coulomb - blockade regime over chaotic cavities to resonant tunneling quantum dots . the second part , sec .
[ sec : boson ] , focusses on quantum dot heat engines that are driven by bosonic degrees of freedom such as phonons , magnons and microwave photons .
we finish with conclusions and an outlook in sec . [
sec : concl ] .
in the following , we discuss different types of heat engines based on quantum dots that are coupled to a hot electronic reservoir . as a first example
, we will analyze a setup based on capacitively coupled quantum dots in the coulomb - blockade regime .
we then continue the discussion with a similar system based on chaotic cavities coupled to reservoirs via quantum point contacts with a large number of open transport channels .
finally , we demonstrate that heat engines based on resonant tunneling through either quantum dots or quantum wells can yield high power in combination with good efficiency . here
, we analyze the thermoelectric properties of two capacitively coupled quantum dots in the coulomb - blockade regime in a three - terminal geometry @xcite .
this model emphasizes the contribution of charge fluctuations , with heat being absorbed merely by means of the coulomb interaction between electrons in the two different conductors .
we demonstrate how a temperature bias across the system can drive a charge current which in turn can be used to generate electrical power .
furthermore , we demonstrate that the device can act as an ideal heat to current converter that can reach carnot efficiency .
the bipartite nature of this setup can be exploited for other purposes such as feedback control @xcite , the detection of dynamical rates @xcite or the investigation of information flows @xcite .
sketch of the coulomb - blockade heat engine .
two single - level quantum dots ( yellow ) are capacitively coupled to each other . the lower gate quantum dot is connected to a single hot reservoir ( red ) .
the upper conductor dot is coupled to two cold electronic reservoirs ( blue ) . ]
we consider the setup shown schematically in .
it consists of two quantum dots @xmath6 in the coulomb - blockade regime .
each dot hosts a single level with energy @xmath7 .
assuming strong onsite coulomb interaction , the dots can host either @xmath8 or @xmath9 excess electron . in the following
, we neglect the electron spin as it will only lead to a renormalization of the tunnel couplings introduced below .
the two quantum dots are capacitively coupled to each other such that they can exchange energy but no particles . the strength of the coupling is characterized by the coulomb energy @xmath10 that is needed to occupy both quantum dots at the same time and will parametrize the energy exchange between the two dots .
the conductor dot , @xmath11 , is tunnel coupled to two electronic reservoirs @xmath12 .
the reservoirs are in local thermal equilibrium and characterized by chemical potentials @xmath13 and temperature @xmath14 .
the gate dot , @xmath15 , is coupled to a single reservoir @xmath16 with chemical potential @xmath17 and temperature @xmath18 . in order to obtain a finite thermoelectric response through the conductor dot to a temperature difference @xmath19 , we need to have energy - dependent tunnel coupling strengths . here , we model this energy dependence by choosing the tunnel couplings @xmath20 between the dot and its respective reservoir @xmath21 to depend on the number of electrons , @xmath22 , on the _ other _ quantum dot .
there are different ways how to realize this energy dependence in an actual experiment .
first of all , the transmission through a tunnel barrier generically depends on energy of the tunneling particle .
this effect is most pronounced for energies close to the barrier height .
second , quantum dots with an excited state that couples asymmetrically to the two leads and decays quickly into the ground state can mimic energy - dependent tunneling rates as well .
finally , additional quantum dots can be added between the conductor dot and its two reservoirs .
these additional dots serve as energy filters that only transmit electrons at a given energy and thereby allow to achieve an effectively energy - dependent tunnel coupling . in order to describe the system , we trace out the noninteracting electronic reservoirs .
the remaining strongly interacting quantum dot degrees of freedom are then characterized by a reduced density matrix for the double quantum dot .
the time evolution of the system is determined by a master equation @xmath23 for the reduced density matrix elements @xmath24 describing the probability to find the double dot empty @xmath25 , occupied with one electron on the conductor @xmath26 or gate dot @xmath27 and doubly occupied @xmath28 . in the sequential tunneling regime
@xmath29 the transition rates @xmath30 follow from fermi s golden rule and are given by @xmath31 the transition rates @xmath32 describe tunneling of electrons onto or off the dot through barrier @xmath33 when the other dot has @xmath22 electrons .
specifically , they are written as @xmath34 where @xmath35 + 1\}^{-1}$ ] denotes the fermi function with temperature @xmath36 and bias voltage @xmath37 . in order to calculate the charge and heat currents as well as their fluctuations and correlations
, we use a full - counting statistics approach . to this end
, we multiply the transition rates with a factor @xmath38 where @xmath39 denotes the charge transferred through barrier @xmath33 in the associated tunneling event . the cumulant generating function @xmath40 is then given by the eigenvalue of the resulting matrix @xmath41 that goes to zero as @xmath42 .
the charge currents are simply given by the derivative of the cumulant generating function with respect to the counting fields , @xmath43 .
similarly , for the energy currents , we introduce counting factors @xmath44 where @xmath45 denotes the energy transferred through barrier @xmath33 in the corresponding tunneling event .
higher - order derivatives give the correlations that will be discussed in section [ sec : cbfluct ] .
finally , heat currents follow from charge and energy currents as @xmath46 .
we remark that while charge and energy currents are conserved , @xmath47 , heat currents in general are not conserved due to joule heating .
current at zero bias voltage in units of @xmath48 as a function of the two level positions @xmath49 and @xmath50 .
parameters are @xmath51 , @xmath52 , @xmath53 and @xmath54 . ] in the following , we want to drive a charge current through the conductor dot simply by applying a temperature difference between the reservoirs of conductor and gate dot , i.e. without applying a bias voltage across the conductor dot . in order to achieve this goal
, two requirements have to be met .
first of all , the left - right symmetry of the system has to be broken .
this can be achieved by choosing the tunnel couplings of the conductor dot to the left and right reservoirs different .
in addition , we also have to break the particle - hole symmetry underlying the system .
this is achieved by having energy - dependent tunneling rates for the conductor dot . in the absence of a bias voltage
, the charge current @xmath2 through the conductor dot can be simply related to the heat current @xmath1 flowing out of the gate dot , @xcite @xmath55 the ratio of heat and charge currents is determined by the ratio of electron charge and charging energy .
furthermore , it depends on the ratio of the different tunnel couplings that enter the problem .
the direction of the charge current can be controlled via the asymmetry of tunnel couplings as well as by the direction of the heat current , i.e. by the sign of the temperature bias @xmath19 .
the largest current , @xmath56 , can be achieved in an optimal configuration when the conductor dot couples only to , say , the left lead when the gate dot is empty and only to the right lead , when the gate dot is occupied .
the heat - driven current as a function of the level positions of the two dots is shown in . as a function of the gate dot level , @xmath50 ,
the current exhibits a peak at @xmath57 with a width given by the temperature of the hot bath , @xmath18 . as a function of the conductor dot level , @xmath49
, the current has a plateau between @xmath58 and @xmath59 .
the borders of the plateau are smeared by the temperature of the cold bath , @xmath18 .
the mechanism giving rise to the heat - driven charge current is the following .
initially , the double dot is completely empty . in a first step , an electron tunnels onto the conductor dot from , say , the left reservoir .
next , an electron tunnels onto the gate dot .
this requires the charging energy @xmath10 which is extracted from the hot reservoir .
afterwards , the electron from the conductor dot tunnels into the right lead , thereby releasing the charging energy into the cold reservoir . in a final step ,
the gate dot is emptied such that the system returns to its initial state .
in one such transport cycle , one quantum of energy , i.e. the charging energy , is transferred from the hot to the cold , giving rise to correlations of the charge and heat currents .
we remark that , in the optimal configuration , one electron is transmitted from the left to the right reservoir for _ every _ absorbed quantum of heat .
this is known as the tight - coupling limit .
( a ) power @xmath60 ( solid lines ) and efficiency @xmath61 ( dashed lines ) as a function of the applied bias voltage for different temperatures of the cold reservoir .
( b ) efficiency at maximum power as a function of the carnot efficiency .
parameters are @xmath51 , @xmath52 , @xmath62 , @xmath63.,title="fig : " ] ( a ) power @xmath60 ( solid lines ) and efficiency @xmath61 ( dashed lines ) as a function of the applied bias voltage for different temperatures of the cold reservoir .
( b ) efficiency at maximum power as a function of the carnot efficiency .
parameters are @xmath51 , @xmath52 , @xmath62 , @xmath63.,title="fig : " ] so far , we demonstrated that our quantum - dot heat engine can convert a heat current into a directed charge current . in a next step ,
we want to generate a finite output power by adding a load to the system .
hence , we apply a bias voltage @xmath4 across the conductor dot against which the heat - driven current can perform work .
the output power is then simply given as @xmath3 .
it is shown as a function of the applied bias in a ) . for zero bias ,
the output power obviously vanishes .
similarly , it vanishes at the so called stopping voltage @xmath64 where the heat- and bias - driven currents compensate each other such that @xmath65 . in between
, the system reaches its maximal output power . for situations close to equilibrium
, the maximum power is reached at half the stopping voltage . as the system
is taken far away from equilibrium by lowering @xmath14 , the point of maximum power is shifted towards larger bias voltages .
another important quantity to characterize the thermoelectric performance of a heat engine is its efficiency of heat to work conversion .
it is defined as the ratio between the output power and the input heat which in our case is given by the heat current flowing out of the hot reservoir , i.e. , we have @xmath5 . in the tight - coupling limit , the efficiency grows linearly with the applied bias voltage and we simply have @xmath66 , cf .
hence , at the stopping voltage @xmath67 the device reaches carnot efficiency @xmath68 indicating that it operates as an optimal heat to current converter . however , at this point , the heat engine operates reversibly and , therefore , does not produce any output power . a more relevant quantity to consider is the efficiency at maximum power .
it is shown as a function of the temperature bias in b ) .
for a small temperature bias , it increases linearly as @xmath69 in agreement with a general thermodynamic bound for systems with time - reversal symmetry @xcite . in the nonlinear regime
, it grows faster than @xmath69 and even reaches @xmath70 for @xmath71 .
we remark , however , that at this point our master equation approach is no longer valid since higher order tunneling contributions that give rise to a broadening of the dot level become important . up to now
, we only discussed the average transport of heat and charge through the quantum - dot heat engine .
however , at the nanoscale , fluctuations of average quantities turn out to be important @xcite .
indeed , we discussed above how the generation of current depends on the correlation of charge fluctuations in the two dots . a way to characterize these fluctuations is to look at the full counting statistics @xmath72 which addresses the probability that @xmath73 electrons have passed through the system in a given time @xmath74 @xcite . in a quantum dot system , this quantity can actually be measured by coupling a charge sensor such as a quantum point contact or a single - electron transistor to the quantum dot @xcite . as the number of electrons on the dot changes , the electrical potential felt by the charge sensor changes , thus leading to a change in the current through it .
this allows to measure the charge on the quantum dot in real time .
interestingly , in the quantum - dot heat engine we can not only access the full counting statistics of charge but also of heat @xcite . measuring the charge of both , the conductor and the gate dot , e.g. , via a quantum point contact that couples asymmetrically to the two dots , allows us to reconstruct the counting statistics of heat from the counting statistics of charge due to the intimate relationship between heat and charge transfer in the system . this way ,
non - equilibrium fluctuation relations can be measured that relate the charge and heat currents @xcite in terms of electron counting .
this is important because in the presence of temperature gradients , fluctuation relations for charge currents @xcite become configuration - dependent @xcite unless one also considers energy currents .
notably , the charge fluctuation theorem becomes universal ( only depending on the thermodynamic forces ) in the tight - coupling limit @xcite .
additional insight of the thermoelectric performance of the quantum - dot heat engine can be accessed by investigating the charge and heat current - current correlations @xcite .
the interest on electronic heat noise has appeared only very recently @xcite .
already in the linear regime the charge - heat cross - correlations are related to seebeck - like and peltier - like coefficients by means of an extended three - terminal fluctuation - dissipation theorem .
far from equilibrium , dimensionless quantities for the auto- and cross - correlations ( charge and heat fano factors , and cross - correlation coefficient ) can be defined .
divergences of the charge fano factor can be used to measure the non - linear thermovoltage .
most interestingly , the charge - heat cross - correlations are maximal in the tight - coupling limit @xcite , leading to carnot - efficient configurations @xcite .
as we have just discussed , capacitively coupled quantum dots in the coulomb - blockade regime can operate as optimal heat to charge current converters that can reach carnot efficiency . yet
, they are of limited practical use for energy harvesting applications as they provide only small currents and output power .
this is a consequence of the fact that transport in these systems occurs via the tunneling of single electrons . in the following
we elucidate the thermoelectric performance of a device based on chaotic cavities coupled to electronic reservoirs via quantum point contacts with a large number of open transport channels @xcite .
our main aim is to discuss how current , power and efficiency behave as the number of open channels is changed .
schematic sketch of the heat engine based on chaotic cavities .
the two cavities are capacitively coupled to each other . in addition , the rectifying cavity is connected to two cold electronic reservoirs at temperature @xmath75 via quantum point contacts .
similarly , the second cavity is connected to a single hot reservoir at temperature @xmath76 .
( reprinted with permission from @xcite .
copyright 2012 american physical society . ) ] we consider two open quantum dots @xmath77 coupled via a mutual capacitance @xmath78 .
each cavity is connected to an electronic reservoir @xmath12 via a quantum point contact , cf . .
the reservoirs are in local thermal equilibrium and described by a fermi function @xmath79 + 1\}^{-1}$ ] with chemical potential @xmath13 and temperature @xmath36 .
interaction effects are captured by capacitive couplings @xmath80 between cavity @xmath81 and reservoir @xmath82 that screen potential fluctuations .
we focus on the semiclassical regime where the number of open transport channels @xmath83 of the quantum point contacts is large .
we assume that dephasing destroys phase information but preserves energy .
in this situation , the cavities are characterized by distribution functions @xmath84 that depend only on energy . for later convenience ,
we write these distributions as @xmath85 the first term corresponds to the average of the reservoir distribution functions weighted with the transmission @xmath86 of the corresponding qpc .
the second term describes fluctuations of the distribution function that have to be determined in the following .
in addition , the cavities are characterized by their potential @xmath87 and associated fluctuations @xmath88 . in order to obtain a finite thermoelectric response
, we need to have energy - dependent transmissions @xmath89 . here , we model this energy dependence as @xmath90 .
while the first term , @xmath91 describes the energy - independent transmissions that depend linearly on the number @xmath83 of open transport channels , the second term captures changes in the transmission due to fluctuations of the cavity potential .
we remark that the energy - dependent term @xmath92 is independent of the number of open transport channels .
the distribution functions of the cavities obey a kinetic equation of the form ( cf .
, e.g. , ref .
@xcite ) @xmath93 where @xmath94 denotes the density of states at the fermi energy in cavity @xmath81 .
the kinetic equation describes how the charge in a given energy interval changes due to changes of the cavity potential @xmath87 , in- and outgoing electron currents through the quantum point contacts as well as due to fluctuations of these currents @xmath95 . here
, the index @xmath96 indicates that we have to sum over all contacts @xmath82 of a given cavity @xmath81 . in a next step ,
we obtain a relation between the fluctuations of the cavity distributions @xmath97 and potentials @xmath88 by expressing the charge inside each cavity in terms of an integral over the distribution functions as well as in terms of the various capacitances and potentials . using this relation between @xmath97 and @xmath88
, we can transform the kinetic equation into a langevin equation for the potential fluctuations @xmath88 .
due to the nonlinearity introduced into the problem by the energy - dependent transmission , the langevin equation has a multiplicative noise term that leads to the it - stratonovich problem in the interpretation of the stochastic integral when converting the langevin equation into a fokker - planck equation @xcite . for the problem at hands
, it turns out that only the kinetic prescription by klimontovich @xcite provides a meaningful solution that exhibits vanishing heat and charge currents in global equilibrium . from the fokker - planck equation , we can obtain the expectation values of the potential fluctuations , @xmath98 and @xmath99 , which subsequently allow us to evaluate the charge current between cavity 1 and its contact @xmath82 via the standard scattering matrix expression @xmath100 we first focus on the situation where a finite temperature bias @xmath101 is applied across the system while there is no bias voltage applied across cavity 1 , i.e. @xmath102 . for the charge current through cavity 1 we find to lowest order in the energy - dependent transmission the compact expression @xmath103 the charge current depends on the asymmetry parameter @xmath104 with @xmath105 , @xmath106 and @xmath107 .
it characterizes both the breaking of left - right symmetry as well as the breaking of particle - hole symmetry due to energy - dependent transmissions . from eq .
, we conclude that in order to have a finite charge current driven by the temperature bias , we need to break both symmetries simultaneously .
we remark that a similar antisymmetric combination of energy - dependent transmissions was found in the expression for the charge current through coulomb - blockaded dots , , as there we also have to break left - right and particle - hole symmetry at the same time .
in addition , the current also depends on the @xmath108 time of the cavity , @xmath109 . here
, @xmath110 denotes the effective conductance of the double cavity while @xmath111 is an effective capacitance that characterizes the coupling between the two cavities . for weakly coupled cavities
, it grows as @xmath112 .
we now turn to the discussion of how the current depends on the number of open transport channels .
as the energy - dependent part of the transmission does not scale with @xmath83 , we find that the current is independent of the number of open channels . for realistic parameters @xcite with @xmath113{ff}$ ] , @xmath114{(mv)}^{-1}$ ] and a temperature bias of
, we obtain @xmath115{na}$ ] .
hence , the current for the cavity heat engine is about two orders of magnitude larger than for the heat engine operating in the coulomb - blockade regime .
the reason for this enhanced current is simply the difference between transport through a tunnel barrier and a fully open quantum channel . in order to generate a finite output power
, we have to apply a bias voltage @xmath116 against which the heat - driven charge current through cavity 1 can perform work .
the output power is then simply given by @xmath117 .
it vanishes when no bias voltage is applied .
similarly , it also vanishes at the stopping voltage @xmath118 ( here , @xmath119 denotes the conductance of cavity 1 ) where heat- and bias - driven charge currents exactly compensate each other such that @xmath120 . the maximal power is generated at half the stopping voltage and is given by @xmath121 ^ 2.\ ] ] surprisingly , the output power drops inversely as the number of open transport channels .
the reason for this lies in the fact that the energy - dependent transmission is independent of the number of open transport channels .
while the heat - driven current thus is independent of the channel number as well , the bias driven current linearly scales with the channel number . hence , the more transport channels are open , the smaller the stopping voltage and , hence , the smaller the output power that can be achieved .
we now turn to the discussion of the efficiency @xmath61 given by the ratio of output power to input heat .
the input heat is given by the heat current flowing between the cavities .
to leading order in the nonlinearity , it is given by @xmath122 i.e. there is a finite heat current even in the absence of energy - dependent transmissions . as the heat current is independent of the applied bias voltage ,
we find that the maximal efficiency and the efficiency at maximum power coincide and are given by @xmath123 analyzing the scaling behaviour of the efficiency with the number of transport channels , we find that it drops inversely to the square of the number of open channels
. this faster decrease as compared to the output power is due to the proportionality of the heat current to the number of transport channels . for realistic parameters
, we find that for a few open channels an output power of a few fw can be generated . at the same time
, the efficiency reaches at most a few percent of the carnot efficiency for a device operating a liquid helium temperatures .
schematic dependence of the output power @xmath60 of a heat engine on its conductance @xmath124 . in the coulomb - blockade regime , power grows linearly with conductance until the coulomb - blockade regime is left . for large conductances
, power decays as the inverse conductance . in between
, the maximal output power is reached for a heat engine that operates with a single open transport channel . ]
so far , we discussed heat engines based on quantum dots in the coulomb - blockade regime and open quantum dots .
we found that both deliver a rather small power . in the coulomb - blockade regime
, power grows linearly with the conductance but is limited by the fact that transport occurs via the tunneling of single electrons .
for chaotic cavities , the power drops inversely with the conductance because the relative importance of the energy - dependent transmission goes down as more and more transport channels open up , cf . the schematic sketch in .
the maximal output power can therefore be expected for a heat engine based on transport through a single quantum channel . a paradigmatic realization of such single channel transport is given by resonant tunneling through quantum dots that we analyze below @xcite .
similar setups have been considered in their dual role as an electronic refrigerator @xcite and successfully been used to cool a micrometer - sized island from to @xcite .
heat engine based on resonant tunneling through quantum dots . a central cavity ( red ) at temperature @xmath125
is coupled via quantum dots to electronic reservoirs ( blue ) at temperature @xmath126 .
light areas in the cavity and reservoirs indicate the thermal broadening of the fermi distribution .
( reprinted with permission from @xcite .
copyright 2013 american physical society . ) ] we consider a setup consisting of a central cavity connected to two electrodes via quantum dots , cf . .
the electrodes are assumed to be in local thermal equilibrium characterized by a fermi distribution with temperature @xmath126 and chemical potential @xmath127 .
each quantum dot has a single resonant level relevant for transport .
the levels are characterized by their width @xmath128 which we assume to be the same for both dots in the following and their level position @xmath129 .
the difference of level positions @xmath130 characterizes the energy an electron can gain in passing through the cavity .
it is different from the level spacing within each dot and can be tuned by applying a gate voltage .
the central cavity is in thermal contact with a hot bath .
this coupling is treated as a third terminal that injects a heat current @xmath1 but no charge in the conductor , as in . for our purposes here , the nature of the hot source needs not to be further specified .
in contrast to the previous model , where the cavity was assumed out of equilibrium , here we make the simple assumption that fast relaxation processes via electron - electron and electron - phonon scattering give rise to a fermi distribution of electrons inside the cavity with temperature @xmath125 and chemical potential @xmath131 .
the cavity temperature and chemical potential are determined by the conservation of charge and energy flowing into the cavity @xmath132 and @xmath133 where @xmath134\ ] ] denotes the charge current flowing between the cavity and reservoir @xmath135 while @xmath136\ ] ] is the energy current flowing between the cavity and reservoir @xmath137 . in the above expressions for the currents
@xmath138 denotes the transmission function of the quantum dot levels @xcite .
it is a lorentzian with width @xmath128 centered around the level position @xmath129 .
the charge conservation condition can be satisfied by placing the cavity potential symmetrically between the resonant levels of the quantum dots @xmath139 and putting the chemical potentials of the reservoirs symmetrically with respect to @xmath131 , @xmath140 .
we first discuss the regime @xmath141 in which an analytical solution of the problem can be obtained that provides us with an intuitive understanding of the underlying physics .
afterwards , we will discuss the regime @xmath142 which we numerically find to yield the largest output power . with narrow resonances , an electron that enters the cavity from the left lead with energy @xmath143 has to gain the precise amount of energy @xmath144 in order to be able to leave the cavity into the right lead . in the steady state ,
rectified charge current @xmath145 and heat current @xmath1 are therefore proportional to each other with the proportionality constant being given by the ratio between electron charge and difference of level positions , @xmath146 again defining a tight - coupling limit .
we now assume that no bias voltage is applied between the two electrodes .
we inject a heat current @xmath1 in such a way as to keep the cavity at a given temperature @xmath125 . as a result
, we find that the charge current @xmath147,\ ] ] flows in response to the temperature difference between the hot cavity and the cold electrodes .
the above expression is valid in the limit where @xmath148 . from
we conclude that the current and , hence , the output power grow both with the level width as well as with the difference of level positions until these quantities exceed the temperature .
the output power that can be generated against an externally applied bias voltage @xmath149 is given by @xmath150 for temperatures larger than @xmath151 and @xmath144 .
here , @xmath152 denotes the stopping voltage at which heat- and bias - driven current compensate each other . at the stopping voltage ,
the device operates reversibly and reaches carnot efficiency . at half the stopping voltage ,
the output power becomes maximal with @xmath153 at this point , the efficiency of heat to work conversion is given by @xmath69 in agreement with general thermodynamic bounds for time - reversal symmetric systems @xcite .
( a ) maximal power as a function of the energy splitting @xmath144 for @xmath154 and level position and width optimized for maximum power .
( b ) maximum power and ( c ) efficiency at maximum power as a function of @xmath155 .
the corresponding optimal values of @xmath156 , @xmath128 and @xmath157 are shown in ( d ) .
( reprinted with permission from @xcite .
copyright 2013 american physical society . ) ] we now turn to the situation of arbitrary level width and position . in this case , the integrals in and have to be evaluated numerically . the main results of our analysis are summarized in where we defined the average temperature @xmath158 and the temperature bias @xmath159 .
( a ) shows the output power as a function of the energy difference @xmath144 . for small @xmath144
the power grows quadratically with @xmath144 in agreement with the analytical result . after reaching a maximum at @xmath160
it decays exponentially for large @xmath144 due to the small number of electrons available at very high energies .
( b ) shows the maximal output power we can obtain when optimizing bias voltage , level positions and level width at the same time .
the corresponding optimal parameters are depicted in ( d ) . while the optimal bias voltage grows linearly with the applied temperature bias , the optimal level position and width are nearly independent of it and given by @xmath160 and @xmath161 .
the maximal power is given by @xmath162 which amounts to about for temperature bias of .
at the same time , the efficiency at maximum power is nearly independent of the temperature bias and given by about @xmath163 .
compared to the coulomb - blockade regime , we therefore loose a factor of two in efficiency .
this is , however , more than compensated by the two orders of magnitude gain in output power .
the reason for this dramatic increase in power is the combination of a strong energy dependence of the transmission functions in combination with a large number of electrons that can pass through a fully open quantum channel .
we estimate that an area of @xmath164{cm^2}$ ] covered with nanoengines that each have an area of @xmath165{nm^2}$ ] can produce an output power of for a temperature bias of . swiss - cheese sandwich heat engine . a large central cavity ( red )
is connected via layers of self - assembled quantum dots ( yellow ) embedded into an insulating matrix ( transparent ) to two cold electronic reservoirs ( blue ) . ]
such a high packing density can be achieved by a strongly parallelized setup shown in . here , the electrodes are connected to a large central cavity via layers of self - assembled quantum dots embedded into an insulating matrix .
electrons tunnel through the quantum dots like through holes in a slice of swiss cheese driven by the thermal bias .
importantly , the positions of the dots in the two layers do not have to match .
apparently , the swiss - cheese sandwich heat engine outperforms a heat engine based on chaotic cavities discussed above .
the reason for this lies in the fact that here we put many optimized channels in parallel while for open quantum dots there are many channels in parallel out of which only a single one is relevant for thermoelectric purposes .
we remark that the heat engine based on self - assembled quantum dots offers the additional advantage that the irregular nature of the quantum dots layers can help to increase scattering of phonons at the interface and thereby reduce unwanted leakage heat currents between the hot cavity and the cold reservoirs .
output power as a function of the width @xmath166 of the gaussian distribution of level positions . in ( a ) the influence of level width that differ from the optimal value are shown .
similarly , in ( b ) the influence of deviations from the optimal bias voltage are shown .
( reprinted with permission from @xcite .
copyright 2013 american physical society . ) ] so far , we considered the ideal situation where all dots have the same , optimal properties . in a real sample
, there will be fluctuations of level positions from dot to dot . in order to investigate in
how far these imperfections deteriorate the performance of the heat engine , we assume that the level positions are distributed according to a gaussian with width @xmath166 centered around the average level position @xmath129 . the output power as a function of the distribution width
is shown in . as expected the power drops down as the width of the distribution increases .
importantly , even for a spread of 10% the power decreases only to 90% of its optimal value , i.e. the proposal can tolerate a certain degree of imperfections .
interestingly , for a given degree of fluctuations , choosing nonoptimal values of the level width or bias voltage can even increase the output power . in the previous section we analysed a heat engine based on resonant tunneling through quantum dots that could be scaled up using self - assembled quantum dots in order to deliver a macroscopic output power . in the following ,
we discuss a related proposal in which the quantum dot layers are replaced by quantum wells @xcite .
the possibility to create high thermoelectric figures of merit by using ridged quantum wells has been pointed out @xcite .
furthermore , it was shown that quantum well structures can be used for refrigeration @xcite .
multilayered thermionic devices which have been proposed for refrigeration purposes are similar in design but differ in the role of the resonance @xcite . compared to heat engines based on resonant tunneling through quantum dots , heat engines based on quantum wells
offer a number of potential advantages .
first of all , electrons inside a quantum well have transverse degrees of freedom .
this gives rise to a larger phase space for tunneling electrons .
hence , quantum wells potentially allow for larger currents and output powers .
second , fabricating a homogenous quantum well can be easier than growing a layer of quantum dots with identical properties ( even though we have seen above that certain fluctuations in the dot properties can be tolerated ) .
finally , narrow quantum wells exhibit subband spacings of the order of several hundreds of mev @xcite which makes them ideal candidates for room temperature applications including spin based thermoelectrics @xcite .
another important difference between quantum dots and wells is that the latter transmit any electron with an energy larger than the subband threshold .
therefore , quantum wells are much less efficient energy filters which can potentially degrade their thermoelectric properties .
schematic of the quantum - well heat engine .
a cavity at temperature @xmath18 ( red ) is coupled to electronic reservoirs at temperature @xmath14 ( blue ) .
the coupling is established via quantum wells with threshold energies @xmath129 .
the light shading inside the quantum wells indicates that electrons with any energy larger than the threshold energy can pass through the quantum well .
( adapted with permission from @xcite .
copyright 2013 iop publishing . ) ] we consider a setup similar to the one discussed in section [ ssec : resonant ] .
it consists of a central cavity in local thermal equilibrium with temperature @xmath18 and chemical potential @xmath131 .
it is coupled to cold electronic reservoirs with temperature @xmath14 and chemical potentials @xmath13 via quantum wells , cf . .
we assume that the quantum wells are noninteracting such that charging effects can be neglected .
it is an interesting topic of future research to investigate the influence of interactions on the thermoelectric properties of quantum - well heat engines .
the cavity potential and temperature are determined by the conservation of charge and energy , @xmath167 and @xmath168 , where @xmath169 and @xmath170 denote the charge and energy current flowing between the cavity and reservoir @xmath82 , respectively .
in addition , we have the heat current @xmath1 that is injected into the cavity from a hot thermal bath and later on serves as a drive for the heat engine .
the charge and energy currents can be derived within a scattering matrix approach as @xcite , @xmath171,\ ] ] and @xmath172.\ ] ] in these expressions , @xmath173 denotes the density of states of the two - dimensional electron gas inside the quantum well with an effective electron mass @xmath174 .
@xmath175 is the surface area of the quantum well .
the energies associated with the electron motion in the quantum well plane and perpendicular to it are @xmath176 and @xmath177 , respectively .
the transmission of the quantum wells is given by @xcite @xmath178 ^ 2/4}.\ ] ] with the ( energy - dependent ) coupling strengths @xmath179 and @xmath180 between the quantum well and reservoir @xmath82 or the cavity , respectively .
the energies @xmath181 denote the subband thresholds at which a new transport channel through the quantum well opens up .
for the following discussion , we consider the limit that the quantum wells are only weakly coupled to the reservoirs and the cavity , @xmath182 .
in addition , we restrict ourselves to the case where only the lowest subband is relevant for transport . due to
the large level spacing of narrow quantum wells , this is a reasonable approximation .
a discussion of thermoelectric transport through a single quantum well that takes into account higher subbands as well can be found in ref . @xcite . under the above assumptions
, the transmission function of the quantum wells simplifies to a delta peak , @xmath183 . in this limit
, the integrals in and can be solved analytically and we obtain for the charge and energy currents @xmath184,\ ] ] and @xmath185,\ ] ] respectively . for simplicity
, we denoted the energy of the single relevant subband in each quantum well as @xmath186 .
furthermore , we introduced the integrals @xmath187 and @xmath188 with the dilogarithm @xmath189 . from
we find that the energy current consists of two different contributions .
the first one is proportional to the charge current while the second term breaks this proportionality .
it arises due to the transverse degrees of freedom and is absent in the case of quantum dots with sharp levels , cf .
[ ssec : qdresult ] .
( a ) maximal output power in units of @xmath190 as a function of the threshold energies of the two quantum wells within linear response for a symmetric configuration @xmath191 .
( b ) efficiency at maximum power in units of the carnot efficiency as a function of the threshold energies of the two quantum wells within linear response for a symmetric setup .
( c ) and ( d ) show the same as ( a ) and ( b ) but for an asymmetric device with @xmath192 .
( reprinted with permission from @xcite .
copyright 2013 iop publishing.),title="fig : " ] ( a ) maximal output power in units of @xmath190 as a function of the threshold energies of the two quantum wells within linear response for a symmetric configuration @xmath191 .
( b ) efficiency at maximum power in units of the carnot efficiency as a function of the threshold energies of the two quantum wells within linear response for a symmetric setup .
( c ) and ( d ) show the same as ( a ) and ( b ) but for an asymmetric device with @xmath192 .
( reprinted with permission from @xcite .
copyright 2013 iop publishing.),title="fig : " ] ( a ) maximal output power in units of @xmath190 as a function of the threshold energies of the two quantum wells within linear response for a symmetric configuration @xmath191 .
( b ) efficiency at maximum power in units of the carnot efficiency as a function of the threshold energies of the two quantum wells within linear response for a symmetric setup .
( c ) and ( d ) show the same as ( a ) and ( b ) but for an asymmetric device with @xmath192 .
( reprinted with permission from @xcite .
copyright 2013 iop publishing.),title="fig : " ] ( a ) maximal output power in units of @xmath190 as a function of the threshold energies of the two quantum wells within linear response for a symmetric configuration @xmath191 .
( b ) efficiency at maximum power in units of the carnot efficiency as a function of the threshold energies of the two quantum wells within linear response for a symmetric setup .
( c ) and ( d ) show the same as ( a ) and ( b ) but for an asymmetric device with @xmath192 .
( reprinted with permission from @xcite .
copyright 2013 iop publishing.),title="fig : " ] in the following , we analyze the thermoelectric properties of the quantum - well heat engine in the linear response regime . for simplicity , we assume that both quantum wells are intrinsically symmetric , i.e. we have @xmath193 .
we parametrize the tunnel couplings as @xmath194 and @xmath195 where @xmath196 denotes the total coupling strength while @xmath197 describes the asymmetry between the couplings of the left and right quantum well .
we , furthermore , introduce the average temperature @xmath198 and the temperature difference @xmath199 .
the charge current through the heat engine to linear order in the applied bias voltage @xmath200 and in the applied temperature difference @xmath201 is given by @xmath202,\ ] ] where we introduced the auxiliary functions @xmath203 and @xmath204 from we directly infer that a finite temperature bias @xmath201 can drive a charge current in the absence of an applied bias voltage @xmath151
. the direction of the charge current can be tuned by adjusting the threshold energies @xmath205 and @xmath206 . by applying a bias voltage @xmath151 against the heat - driven current
, the heat engine can perform work .
the resulting output power @xmath207 becomes maximal at half the stopping voltage where it takes the value @xmath208 in order to evaluate the efficiency of heat to work conversion , again given by the ratio between output power and input heat , we need to calculate the heat current injected from the hot bath . at half the stopping voltage
, it takes the form @xmath209 where the function @xmath210 satisfies @xmath211 .
its complete analytical expression can be found in ref .
hence , the efficiency at maximum power is given by @xmath212 in the following , we discuss the output power and efficiency in more detail , starting with a symmetric setup @xmath191 .
in this case , both power and efficiency are symmetric with respect to an interchange of @xmath205 and @xmath206 , cf . .
the output power takes its maximal value @xmath213 when one level is deep below the equilibrium chemical potential while the other is located at approximately @xmath214 above it .
an explanation for this behaviour will be given below .
the efficiency at maximum power takes its maximal value @xmath215 when both levels are above the equilibrium chemical potential and satisfy @xmath216 .
however , for these parameters there is only an exponentially suppressed number of electrons that can contribute to transport such that the output power in this regime becomes vanishingly small . for level positions that optimize the output power
, we find an efficiency at maximum power of about @xmath217 .
hence , the quantum - well heat engine is not as efficient as a quantum - dot based setup in the limit of narrow resonances @xmath218 .
the reason for this lies in the different energy - filtering properties of quantum dots and wells .
quantum dots with narrow resonances transmit energies only at a single energy .
hence , they reach the tight - coupling limit where heat and charge current are proportional to each other . in this situation , the efficiency at maximum power is then given by @xmath219 .
quantum wells on the other hand transmit any electron with an energy larger than the threshold voltage as in this case the energy @xmath220 can be decomposed into a part associated with the motion in the plane of the well and perpendicular to it , @xmath221 .
hence , high - energy electrons can be transmitted through the well if most of their energy is in the perpendicular degrees of freedom such that @xmath176 matches the resonance condition . as a result
, quantum wells act as much less efficient energy filters .
given the rather weak energy filtering properties of quantum wells , it is quite surprising that the efficiency at maximum power is only a factor of three smaller than for quantum dots with level widths of the order of @xmath222 the configuration that gives the largest output power , as discussed in sec .
[ ssec : qdresult ] . to understand this feature
, we consider the situation depicted in .
the right quantum well has a threshold energy slightly above the equilibrium chemical potential .
as the number of electrons with energies much larger than @xmath206 is exponentially small , it acts as a good energy filter .
for the left well , the energy filtering relies on a different mechanism .
electrons with energy @xmath220 can enter the cavity only if the associated state in the left reservoir is occupied , @xmath223 and , at the same time , the corresponding state in the cavity is empty , @xmath224 .
this defines an energy window of about @xmath222 in which electrons are transmitted through the well .
this explains why both quantum - dot and quantum - well based heat engines have similar efficiencies .
we now turn to the asymmetric case @xmath225 . in this situation ,
the output power and efficiency are no longer symmetric with respect to an interchange of @xmath205 and @xmath206 .
in fact , as can be seen in c ) and d ) , we find that power and efficiency are strongly suppressed if @xmath226 and @xmath227 for an asymmetry @xmath228 ( the roles of @xmath205 and @xmath206 are interchanged for @xmath229 ) .
however , for @xmath230 and @xmath231 we find that the power can be enhanced by up to @xmath232 for an asymmetry of about @xmath233 while the efficiency at maximum power is even nearly doubled compared to the symmetric case .
we finally estimate what output power can be expected for a realistic device . for a gaas based structure with an effective electron mass @xmath234 , level width of @xmath235 and asymmetry
@xmath192 operating at room - temperature @xmath236{k}$ ] , we obtain a maximal output power of @xmath237{w cm^{-2}}$ ] for a temperature bias of @xmath238{k}$ ] .
hence , the quantum - well heat engine is about a factor of two more powerful than the previously discussed quantum - dot heat engine .
we remark that materials with higher effective mass can yield even larger output powers . similarly to the quantum - dot case , we also find that a quantum - well heat engine is robust with respect to fluctuations of the threshold energies .
as can be seen in c ) the output power is hardly affected at all by fluctuations of the right threshold energy as long as @xmath239 is fulfilled .
fluctuations of the left threshold energy are more crucial , but even here we find that variations of as much as @xmath222 reduce the output power by only 20% .
( a ) maximized output power ( red ) and efficiency at maximum power ( blue ) as a function the temperature bias @xmath201 .
( b ) parameters that maximize the output power .
( reprinted with permission from @xcite .
copyright 2013 iop publishing.),title="fig : " ] ( a ) maximized output power ( red ) and efficiency at maximum power ( blue ) as a function the temperature bias @xmath201 .
( b ) parameters that maximize the output power .
( reprinted with permission from @xcite .
copyright 2013 iop publishing.),title="fig : " ] we now turn to a discussion of the thermoelectric performance of our heat engine in the nonlinear regime .
similarly , to the quantum dot heat engine discussed in sec .
[ ssec : qdresult ] , we optimize the applied bias voltage , the threshold energies of the wells as well as the asymmetry @xmath240 for a given temperature bias @xmath201 . the corresponding results are shown in b ) .
the optimal bias voltage grows linearly with the temperature bias .
the optimal asymmetry @xmath241 is independent of @xmath201 . the optimal threshold energy for the right quantum well decreases only slightly with the temperature bias while the threshold energy of the left well should be chosen as @xmath242 independent of @xmath201 .
the optimized power plotted in a ) grows quadratically with the applied temperature difference and is approximately given by @xmath243 . due to the quadratic dependence on @xmath201 we obtain the same output power for a given value of @xmath201 in the linear and nonlinear regime
however , as the efficiency at maximum power grows linearly with the applied temperature bias , the device should be operated as much in the nonlinear regime as possible . in the extreme case of @xmath244 ,
the quantum - well heat engine reaches an efficiency at maximum power of @xmath245 , i.e. it is about as efficient as the quantum - dot heat engine but delivers twice the power .
so far , we discussed three - terminal heat engines based on electronic degrees of freedom only . in the following
, we turn to a different class of energy harvesters where heat is injected from a third _ bosonic _ reservoir .
first , we will discuss several heat engines that are powered by phonons @xcite .
we will then analyze a magnon - driven heat engine that offers an additional spin degree of freedom , thereby establishing the connection to spin caloritronics @xcite .
finally , we will discuss heat engines driven by photons .
these photons can either be absorbed from the electromagnetic environment @xcite or be specifically emitted by the hot source into a resonant superconducting microwave cavity that serves as a quantum bus for heat exchanged between spatially separated mesoscopic conductors @xcite .
( a ) a quantum dot ( brown ) is coupled to two electronic reservoirs ( pink ) with temperatures @xmath246 and @xmath247 as well as to a phonon bath ( green ) with temperature @xmath248 .
( reprinted with permission from @xcite .
copyright 2010 american physical society . )
( b ) aharonov - bohm ring with a quantum dot embedded in one of the arms .
( reprinted with permission from @xcite .
copyright 2012 american physical society.),title="fig : " ] ( a ) a quantum dot ( brown ) is coupled to two electronic reservoirs ( pink ) with temperatures @xmath246 and @xmath247 as well as to a phonon bath ( green ) with temperature @xmath248 .
( reprinted with permission from @xcite .
copyright 2010 american physical society . )
( b ) aharonov - bohm ring with a quantum dot embedded in one of the arms .
( reprinted with permission from @xcite .
copyright 2012 american physical society.),title="fig : " ] as a first example of a boson - driven heat engine , we consider a single - level quantum dot coupled to two electronic reservoirs as well as a to a local phonon mode , cf .
a ) @xcite .
experimentally , such a system can be realized in a single - molecule junction where generically the electron localized on the molecule couples strongly to the molecular vibrations .
this electron - phonon coupling gives rise to a number of interesting phenomena such as franck - condon blockade of transport at small bias voltages @xcite , nonequilibrium phonon distributions @xcite , lasing of optical phonons @xcite , two - terminal thermoelectric effects @xcite and allows the study of fluctuation theorems in the presence of fermionic and bosonic degrees of freedom @xcite .
here , we demonstrate that such a setup can also serve as a heat engine driven by the temperature bias between the electronic and phononic reservoirs @xcite .
the hamiltonian of the system can be written as the sum of three parts , @xmath249 the first term describes the left , @xmath250 , and right , @xmath251 , electronic reservoir in terms of noninteracting , spinless electrons ( we neglect the spin degree of freedom in the following as it only leads to a trivial renormalization of the tunnel couplings defined below ) @xmath252 the two electronic reservoirs are at temperature @xmath36 and chemical potential @xmath13 , respectively . for later convenience , we parametrize the temperatures and chemical potentials as @xmath253 and @xmath254 . the molecule inside the junction is described by the hamiltonian @xmath255 the molecule hosts a single level relevant for transport with energy @xmath256 .
it is coupled to a single dispersionless vibrational mode with energy @xmath257 via coupling strength @xmath128 . while in principle
the phonon distribution could be a nonequilibrium one , in the following we consider the situation where it is thermal with temperature @xmath258 due to the coupling to the environment .
finally , tunneling between the dot and the electrodes is given by @xmath259 where the tunnel matrix elements @xmath260 are related to the tunnel coupling strength as @xmath261 .
for the system at hands , the charge current @xmath2 through the molecule as well as the electronic and phononic heat currents , @xmath262 and @xmath263 , can be evaluated using a nonequilibrium green s function approach that takes into account the tunneling between the dot and the leads exactly and performs a perturbative expansion in the electron - phonon coupling @xmath128 up to second order . in linear response , the currents are related to the corresponding thermodynamic forces @xmath264 , @xmath155 and @xmath265 via the onsager matrix @xmath266 the full analytic expressions for the onsager coefficients can be found in ref . @xcite . while the coefficients @xmath124 , @xmath267 and @xmath268 contain contributions that are both due to elastic and inelastic transport through the molecule , the other onsager coefficients result from inelastic processes only .
the coefficient @xmath269 describes a charge current response to a temperature difference between phonons and electrons . in order to have a finite @xmath269 ,
both the left - right as well as the particle hole symmetry need to be broken , i.e. @xmath270 must be energy - dependent and fulfill @xmath271 . similarly , the electronic heat current due to a temperature difference between electrons and phonons , characterized by the onsager coefficient @xmath272 , requires a breaking of left - right and particle - hole symmetry .
an extension of the setup we just discussed consists of an aharonov - bohm geometry with a molecular quantum dot embedded in one of the arms @xcite , cf .
b ) . transport through the reference arm of the interferometer is described by the new contribution to the hamiltonian @xmath273 where @xmath274 denotes the aharonov - bohm flux through the ring .
due to the aharonov - bohm flux , the onsager matrix becomes @xmath274-dependent , @xmath275 , such that the onsager relations @xcite are satisfied .
the analytic expressions for the onsager coefficients that are given in ref .
@xcite exhibit three different types of flux dependence .
first , there are contributions proportional to @xmath276 that arise from interference contributions .
second , there are terms proportional to @xmath277 due to contributions from time - reversed paths .
third , there are @xmath278 contributions that emerge from the coupling to the phonons .
while the former two contributions are even in the flux , the latter one is odd .
the odd contributions in the off - diagonal onsager coefficients can potentially lead to a large efficiency at maximum power @xcite that overcomes the linear response limit @xmath219 that exists for time - reversal symmetric systems @xcite .
however , for the system at hands , these efficiency bounds have not yet been investigated .
the idea of phonon - driven thermoelectric transport in multi - terminal devices was further discussed in the context of phonon - assisted hopping @xcite .
furthermore , a p - i - n - diode structure that drives a charge current by harvesting energy from a hot phonon source was discussed in ref .
it was theoretically estimated that such a device based on a bi@xmath279te@xmath280/si superlattice can have a figure of merit larger than 1 when operated at room temperature , thus making it a promising candidate for energy harvesting applications .
we now turn to a quantum - dot heat engine that is driven by spin waves from a ferromagnetic insulator @xcite , cf . .
compared to the phonon - driven setups that we discussed in the previous section , the magnon - driven heat engine offers a number of advantages .
first of all , magnons are potentially easier to control than phonons . as phonons exist in any material , it is hard to avoid leakage heat currents from the hot phonon bath to the cold electronic reservoirs .
magnons , in contrast , exist only in magnetic materials and couple via short - range exchange interactions which facilitates coupling them only to the quantum dot degrees of freedom .
another advantage of the magnon - heat engine is that it does not rely on energy - dependent tunnel couplings . instead , here
the necessary asymmetry between electrons and holes is introduced into the system by the spin - dependence of the tunnel barriers .
finally , the magnon - driven heat engine provides an example of a spin caloritronic heat engine @xcite that allows to drive spin - polarized charge currents as well as pure spin currents by thermal gradients .
alternative spintronic heat engines based on nanowires with domain walls have been proposed in ref .
@xcite .
a related setup has been discussed in ref .
@xcite where a system consisting of a quantum dot coupled to ferromagnetic electrodes with spin waves has been taken into account . keeping the magnons inside the ferromagnets at a different temperature than the electrons , e.g. , by microwave excitation of spin precession , gives rise to spin - polarized charge currents . while this effect was termed magnon - assisted transport in ref .
@xcite in analogy to phonon - assisted tunneling , it can also be viewed as a three - terminal thermoelectric device . schematic sketch of a magnon - driven quantum - dot heat engine .
a single - level quantum dot ( blue ) is tunnel - coupled to two ferromagnetic metals ( yellow ) .
in addition , it is also exchange - coupled to a ferromagnetic insulator ( green ) serving as a source of spin waves .
( reprinted with permission from @xcite .
copyright 2012 iop publishing . ) ] we consider a single - level quantum dot in the coulomb - blockade regime tunnel coupled to two ferromagnetic electrodes @xmath12 at temperature @xmath281 as well as exchange coupled to a ferromagnetic insulator that serves as a source of magnons with temperature @xmath282 .
the ferromagnetic electrodes are modeled in the spirit of a stoner model as a noninteracting electron gas with a constant but spin - dependent density of states @xmath283 .
it is related to the spin polarization @xmath284 that varies between @xmath285 for a normal metal and @xmath286 for a half - metallic ferromagnet . in the following ,
we assume identical polarizations for the two leads , @xmath287 .
the ferromagnetic insulator is modeled as a heisenberg chain of exchange - coupled spins . using a holstein - primakoff transformation ,
it can be described in terms of bosonic operators that create and annihilate magnons . at low temperatures where the average magnon number is small
, the ferromagnetic insulator behaves as a noninteracting magnon gas with dispersion relation @xmath288 .
the quantum dot has a single spin - split level with energy @xmath289 .
the zeeman splitting @xmath290 due to an externally applied magnetic field determines the energy of the magnons that the heat engine can harvest .
the coulomb energy @xmath10 that is required to occupy the dot with two electrons at the same time is assumed to be infinite .
we remark that taking into account a finite value of @xmath10 does not give rise to qualitatively different results .
tunneling between the dot and the ferromagnetic electrodes is characterized by the spin - dependent tunnel couplings @xmath291 . for later convenience
we also introduce the total tunnel coupling strength @xmath292 .
the coupling between the dot and the ferromagnetic insulator is due to an exchange interaction with spectral weight @xmath293 that flips the spin of the dot and emits or absorbs a magnon in the insulator . as in the following we only need the spectral weight evaluated at the zeeman splitting of the dot
, we will omit its energy dependence and write @xmath294 .
transport through the system is described via a standard master equation approach that integrates out the noninteracting fermionic and bosonic degrees of freedom in the reservoirs and characterizes the quantum dot by its reduced density matrix . within this approach ,
the spin - resolved electron and magnon currents can then be evaluated in a straightforward way .
we count particle currents as positive when they flow from the reservoir into the dot .
( a ) spin - resolved electron and magnon currents as a function of the magnon temperature for parallel and antiparallel geometry . parameters
are @xmath295 , @xmath296 , @xmath297 and @xmath298 .
( b ) power and efficiency as a function of applied bias voltage for different polarizations and other parameters as in the left panel .
( reprinted with permission from @xcite .
copyright 2012 iop publishing.),title="fig : " ] ( a ) spin - resolved electron and magnon currents as a function of the magnon temperature for parallel and antiparallel geometry .
parameters are @xmath295 , @xmath296 , @xmath297 and @xmath298 .
( b ) power and efficiency as a function of applied bias voltage for different polarizations and other parameters as in the left panel .
( reprinted with permission from @xcite .
copyright 2012 iop publishing.),title="fig : " ] we now turn to a discussion of the thermoelectric performance of the magnon - driven heat engine .
shows the spin - resolved electron and magnon currents that flow in response to a temperature difference between electrons and magnons . for parallely magnetized electrodes , we find that as soon as @xmath299 , we have finite electron and magnon currents flowing through the dot .
for any temperature bias , we observe that the currents of spin up and spin down electrons have equal magnitude but opposite sign , @xmath300 . hence ,
the total charge current @xmath301 vanishes while the spin current @xmath302 is finite , i.e. a pure spin current is generated .
the physical picture behind this is that spin up electrons tunnel into the dot , absorb a magnon to flip their spin and then tunnel out again . on average , for each spin up electron entering the dot through a given tunnel barrier , there is a spin down electron leaving the dot through the same barrier such that @xmath303 and @xmath304 . for antiparallel magnetizations of the ferromagnetic electrodes , the magnitudes of @xmath305 and @xmath306 are no longer equal to each other such that now a finite , spin - polarized charge current flows through the dot .
it arises as spin up electrons preferably tunnel in from the left electrode whereas spin down electrons preferably tunnel out to the right electrode due to the spin - dependent tunnel couplings . in the following ,
we focus on the antiparallel case and discuss the output power @xmath60 and efficiency @xmath61 of heat to work conversion when an external bias voltage @xmath4 is applied against the thermally driven current .
shows the output power as a function of @xmath4 for different polarizations .
it grows from @xmath307 at vanishing bias to a maximal value and then drops down again to zero at the stopping voltage .
furthermore , it grows as the polarization is increased because the energy filtering properties of the heat engine become more pronounced , reaching the tight - coupling limit for @xmath308 .
the efficiency @xmath61 shown in exhibits qualitatively different behaviour for @xmath309 and @xmath308 . in the former case
, it grows from @xmath310 at @xmath311 to a finite value and goes down to zero at the stopping voltage .
we remark that the maximal efficiency in general occurs for a different bias voltage than the maximal power . for @xmath308 ,
the efficiency just grows linearly with the applied voltage and reaches the carnot efficiency at the stopping voltage as expected in the tight - coupling limit . the corresponding efficiency at maximum power @xmath312 is given by @xmath69 in the linear response regime while it satisfies @xmath313 in the nonlinear regime .
apart from phonons and magnons , one can also make use of photons to drive mesoscopic heat engines .
heat transfer due to photons in nanoscale circuits has been studied theoretically @xcite as well as been experimentally observed @xcite . here , we present two different examples of photon harvesting . in the first case ,
a quantum - dot based setup is used to harvest microwave photons from the electromagnetic environment @xcite . a similar proposal used to harvest energy from visible light
was analyzed in ref .
@xcite . in the second example
, we consider a system of two double quantum dots connected via a superconducting microwave cavity .
the latter serves as a quantum bus for heat flow between the quantum dots .
both types of setup offer the advantage of spatially separating the hot and the cold part of the heat engine .
this potentially reduces leakage heat currents due to substrate phonons and can therefore help to achieve highly efficient heat engines .
( a ) heat engine that harvests energy from the electromagnetic environment . a quantum dot structure
is coupled to two electronic reservoirs at temperature @xmath314 .
the circuit connecting the two leads contains an impedance @xmath315 at the environment temperature @xmath316 .
realizations of the heat engine with a single dot and a double dot are shown in ( b ) and ( c ) , respectively .
( reprinted with permission from @xcite .
copyright 2012 american physical society . ) ] we consider a mesoscopic heat engine as discussed in ref .
@xcite that consists of one or two quantum dots coupled to electronic reservoirs at temperature @xmath314 .
the electronic reservoirs are connected to each other via an external circuit with impedance @xmath315 that is kept at temperature @xmath316 .
we assume that the tunnel coupling between the dots and the leads is weak , such that a lowest order description of tunneling is valid .
we , furthermore , assume that the relaxation time of the electromagnetic environment is much shorter than the average time between subsequent tunneling events .
in this situation , transport through the system can be described by @xmath317 theory @xcite .
the rate for an electron transition from system @xmath81 to @xmath137 is then given by @xmath318 here , @xmath74 denotes the tunnel matrix element for a transition from @xmath81 to @xmath137 .
the density of states for electrons @xmath319 is given by @xmath320 for a quantum dot with discrete levels and by @xmath321 for a metallic island or an electrode where @xmath322 denotes the density of states at the fermi energy .
similarly , the density of states for holes @xmath323 is given by @xmath320 for quantum dots and @xmath324 $ ] for metallic systems .
finally , @xmath325 denotes the probability density that the tunneling electron exchanges the energy @xmath326 with the environment . for a simple tunnel junction without any embedded quantum dot system ,
there is no directed charge current in the absence of an applied bias voltage as the rectification of thermal fluctuations requires the presence of a nonlinearity in the heat engine . the simplest way to achieve such
a nonlinearity is to add a single quantum dot or metallic island into the junction , cf .
b ) . in the limit of strong coulomb blockade ,
the quantum dot or metallic island is either empty or occupied with a single excess electron with energy @xmath220 .
the corresponding occupation probabilities @xmath327 and @xmath328 in the stationary state follow from a simple rate equation @xmath329 with @xmath330 . here
, @xmath331 denotes the transition rate for an electron tunneling left ( right ) through barrier @xmath82 evaluated according to . in the following , we assume that the right tunnel barrier has a much larger transition rate than the left barrier .
we , furthermore , assume that its capacitance is larger than that of the left junction such that it effectively decouples from the environment .
under these conditions , we obtain a directed charge current that is simply given by @xmath332 and thus finite even in the absence of an applied bias voltage .
we remark that the above expression also describes the current through a junction between two quantum dots with energy levels @xmath205 and @xmath206 such that @xmath333 if the appropriate definitions for the electron and hole density of states are inserted in and the dot is preferably singly occupied . by applying a bias voltage @xmath4 against the thermally driven current , we can generate a finite output power .
as for any coulomb - blockade heat engine , the output power is limited by the operation temperature .
for a device operating at a temperature of about , one obtains a power in the femtowatt range .
the efficiency of the heat engine depends strongly on the junction type . for a junction between two quantum dots with discrete energy levels ,
the tight - coupling limit is reached as every electron that is transferred between the dots has to absorb a photon with energy @xmath220 .
consequently , the heat engine can reach carnot efficiency and achieve a large efficiency at maximum power . for junctions between a discrete level and a metal or between two metals
, the situation is different .
now photons of different energies can be absorbed , leading to a significantly lower efficiency .
interestingly , the efficiency depends on whether the environment is hotter or colder than the electronic system . for a cold electron system
, the sharp fermi functions help to achieve unidirectional transport against the bias voltage such that decent efficiencies can be achieved .
in contrast , for a hot electron system the thermally smeared fermi function give rise to a current flow with the bias voltage such that only small efficiencies are possible . as a second example of a photon - driven heat engine we now consider a system where a superconducting microwave cavity connects two mesoscopic conductors , cf . .
such hybrid structures that allow the investigation the interplay of light and matter at the nansocale have recently gained a lot of interest both theoretically @xcite as well as experimentally @xcite in the context of circuit quantum electrodynamics .
similar to the previously discussed energy harvester , the hybrid microwave cavity heat engine @xcite also allows to separate the hot and the cold part of the engine by a macroscopic distance of the order of a centimeter .
this suppresses leakage heat currents far more efficiently than vacuum nanogaps that are only a few nanometer wide @xcite .
in addition , the heat engine has the advantage that the cavity helps in efficiently transferring photons from the hot to the cold side whereas otherwise photons just get randomly emitted into all directions and can be lost for the energy harvesting purpose . ( a ) sketch of the microwave cavity heat engine .
double quantum dots ( yellow ) are coupled to hot ( red ) or cold ( blue ) electronic reservoirs .
the level positions of the dots are tunable via voltages applied to gates ( green ) .
the two double dots are connected via a superconducting microwave cavity ( gray ) .
( b ) operation scheme of the heat engine .
electrons enter the excited state of the hot dqd and leave the ground state after emitting a photon into the cavity . at the cold dot ,
electrons enter the ground state preferrably from the left and leave the excited state to the right after absorbing a photon from the cavity . as a result of this asymmetry , a directed charge current flows through the cold dot .
( reprinted with permission from @xcite .
copyright 2014 american physical society.),title="fig : " ] ( a ) sketch of the microwave cavity heat engine .
double quantum dots ( yellow ) are coupled to hot ( red ) or cold ( blue ) electronic reservoirs .
the level positions of the dots are tunable via voltages applied to gates ( green ) .
the two double dots are connected via a superconducting microwave cavity ( gray ) .
( b ) operation scheme of the heat engine .
electrons enter the excited state of the hot dqd and leave the ground state after emitting a photon into the cavity . at the cold dot ,
electrons enter the ground state preferrably from the left and leave the excited state to the right after absorbing a photon from the cavity . as a result of this asymmetry , a directed charge current flows through the cold dot .
( reprinted with permission from @xcite .
copyright 2014 american physical society.),title="fig : " ] the heat engine consists of two double quantum dots @xmath77 , connected to each other via a microwave cavity . due to
strong coulomb interactions , the double dots can be either empty or occupied with a single electron . in the following ,
we neglect the electron spin as it only renormalizes the tunnel couplings in a trivial way .
the eigenstates of the singly - occupied dot can be expressed as linear combinations of the left @xmath334 and right @xmath335 double quantum dot states as @xmath336 the mixing angle @xmath337 that characterizes the hybridization of the dot levels depends on the level positions and the interdot tunnel coupling .
it can be controlled by gate voltages applied to the double quantum dot .
the energy difference between the eigenstates @xmath338 and @xmath339 can be tuned independently of @xmath337 such that it matches the resonance frequency @xmath340 of the microwave cavity . the coupling between the double dots and the cavity with coupling strength
@xmath341 then induces transitions between the ground and excited state of the dot accompanied by the absorption or emission of a microwave photon in the cavity .
each quantum dot is furthermore tunnel coupled to two electronic reservoirs @xmath342 at temperature @xmath343 and chemical potential @xmath344 with tunnel coupling strength @xmath345 . in the following , we assume for simplicity symmetric tunnel couplings for each double quantum dot , @xmath346 .
we remark that the actual transition rates of the system depend not only on the tunnel coupling @xmath347 but also on the mixing angle @xmath337 that can effectively break the left - right symmetry within a double quantum dot .
we describe transport through the system again within a generalized master equation approach that integrates out the noninteracting electronic reservoirs and characterizes the remaining quantum system consisting of the two double quantum dots and the microwave cavity by its reduced density matrix . in the limit of strong electron - photon coupling , @xmath348
, coherences between different eigenstates of the quantum system can be neglected to lowest order in the tunnel coupling such that a simple rate equation description holds .
it is this limit that we will consider first before taking into account the effects of finite electron - photon coupling as well as of relaxation and dephasing within the double quantum dots . before discussing the thermoelectric performance of the heat engine ,
we elucidate the basic operation principle that drives a charge current through the cold dot in the absence of any applied bias voltage . to this end , we choose a situation where the mixing angle of the hot dot is @xmath349 . in this case , the eigenstates are the bonding and antibonding states . as they do not break left - right symmetry , there is no directed charge current through the double dot . for the cold double dot ,
we choose the mixing angle @xmath350 such that the ground state couples more strongly to one electrode while the excited states couples more strongly to the other one .
electrons then tunnel into the ground states of the cold double dot from one side , absorb a photon from the cavity to get into the excited state and then tunnel out to the other side . as there is a net flow of heat ( and hence microwave photons ) from the hot to the cold double dot , we thus obtain a finite directed charge current through the cold double dot .
we remark that the direction of the charge current depends on the mixing angle as it determines the asymmetry of eigenstates . unlike the energy dependence of tunnel couplings that is required for the heat engines discussed in sec .
[ sec : cbfluct ] , the asymmetry can therefore be changed in a controlled way by manipulating @xmath351 via gate voltages .
thermoelectric performance of the cavity heat engine .
( a ) heat - driven charge current in units of @xmath352 as a function of the temperature bias for different values of the mixing angle @xmath351 .
the analytic results ( dashed lines ) compare well with the full numerics ( solid lines ) .
parameters are @xmath353 and @xmath354 .
( b ) output power in units of @xmath355 and ( c ) efficiency over carnot efficiency as a function of the applied bias voltage .
the analytical results for the power ( dashed lines ) agree well with the numerics ( solid lines ) .
parameters are @xmath356 and @xmath357 .
( d ) efficiency at maximum power and ( e ) maximum power as a function of the temperature bias .
the optimal values of the frequency @xmath257 and voltage @xmath358 are shown in ( f ) .
( reprinted with permission from @xcite .
copyright 2014 american physical society . ) ] the thermoelectric performance of the microwave cavity heat engine is summarized in . in panel
a ) we show that charge current @xmath359 that flows through double quantum dot 2 without an applied bias voltage .
it grows linearly with the applied temperature bias and saturates for @xmath360 . for @xmath361 , the tight - coupling limit
is reached where one electron is transferred through the cold double dot for each photon that is transferred through the cavity . as a consequence , we find that the charge current @xmath359 is proportional to the heat current @xmath362 flowing out of the hot reservoirs , @xmath363 , where the ratio of these two currents is determined only by the electron charge and the photon energy . in the regime where the cavity is mostly empty ,
i.e. , @xmath364 the charge current is approximately given by the analytic expression @xmath365 that agrees well with the full numerical result , cf .
the dashed lines in a ) .
as can be inferred from , the current becomes largest when @xmath366 .
we will therefore focus on this limit in the following discussion .
the output power against an externally applied bias voltage @xmath358 is depicted in b ) .
it shows a typical behaviour by growing from @xmath307 at @xmath367 to a maximal value and dropping to zero at the stopping voltage .
for @xmath368 it is well approximated by @xmath369.}\ ] ] the efficiency in general shows behaviour qualitatively similar to the output power . only in the tight - coupling limit it instead grows linearly with the applied bias voltage and reaches carnot efficiency at the stopping voltage .
the results for the maximum power @xmath370 and the associated efficiency at maximum power @xmath312 together with the optimizing values of @xmath257 and @xmath358 are shown in d)-f ) . in the tight - coupling limit
, @xmath312 grows as @xmath69 in the linear response regime . in the nonlinear regime
, it grows more quickly and reaches @xmath70 for @xmath371 . for @xmath372 , the efficiency at maximum power
is slightly reduced but exhibits a qualitatively similar behaviour .
influence of electronic relaxation and dephasing rates on the heat - driven current . in the optimal case
, the current takes the value @xmath373 .
( reprinted with permission from @xcite .
copyright 2014 american physical society . ) ] in current experiments on circuit quantum electrodynamics relaxation and dephasing within the double quantum dots are the major obstacles in reaching the strong coupling limit .
when taking into account a finite value of the electron - photon coupling @xmath341 together with relaxation and dephasing processes with rates @xmath374 and @xmath375 , respectively , we find that the charge current through the system is reduced to @xmath376 the corresponding current suppression is shown in as a function of @xmath374 and @xmath375 . even when taking these imperfections into account , we estimate that for realistic system parameters one can achieve charge currents of the order of @xmath377{pa}$ ] clearly within the reach of current experiments .
in this review we discussed different types of thermoelectric energy harvesters based on multi - terminal quantum dot and well setups .
we presented nano heat engines based on coulomb - coupled conductors .
in particular , we looked at a setup based on coulomb - blockade quantum dots that turned out to be ideally efficient but only gives small currents and powers .
a similar system based on chaotic cavities was shown to yield much larger currents but had a significantly reduced efficiency .
the optimal heat engine that has both a large output power in addition to a good efficiency was then found to be based on resonant tunneling through quantum dots or quantum wells . in the second part of this review , we discussed different types of heat engines that are powered by absorbing bosons from the environment .
we analyzed heat engines based on molecular junctions where electrons couple to phononic degrees of freedom from which energy is harvested .
we then turned to a related setup that is driven by magnons from a ferromagnetic insulator .
this type of setup allows one to drive pure spin currents as well as spin - polarized charge currents and therefore makes connection to the emerging field of spin caloritronics . finally , we discussed setups that harvest microwave photons from the electromagnetic environment or from a superconducting microwave cavity thus providing a bridge between energy harvesting and circuit quantum electrodynamics .
the field of energy harvesting with mesoscopic conductors of course still faces a number of open questions .
while most of the setups presented in this review were described in terms of minimal models , it would be interesting to have more realistic descriptions that include , e.g. , charging effects in quantum wells or provide a microscopic model of heat injection into the central region of the resonant tunneling heat engines .
furthermore , the influence of phonons has to be better understood as they not only degrade the efficiency but also make it hard to maintain a given temperature bias across the device
. in particular , it would be desirable to investigate systems where an analytic treatment of their influence can be made and to invent setups where they can be controlled in a systematic way .
another interesting question addresses the thermoelectric performance of multi - terminal heat engines with broken time - reversal symmetry .
theoretically , these devices allow to get arbitrarily close to carnot efficiency at finite output power . but how close can any physical realization actually get ? finally , the field of mesoscopic energy harvesting will greatly benefit from experiments that demonstrate that the theoretical ideas presented in this review can also be put into practice .
above all , we are deeply indebted to the support and inspiration of the late markus bttiker during many years of collaborations that led to our works presented in this review .
we furthermore acknowledge numerous fruitful discussions with c. bergenfeldt , h. buhmann , c. flindt , f. hartmann , p. jacquod , a. m. lunde , r. lpez , i. martin , l. molenkamp , m. moskalets , p. samuelsson , d. snchez , j. splettstoesser , h. thierschmann , r. whitney and l. worschech .
we also acknowledge financial support from the european strep project nanopower , the swiss nsf via the nccr qsit , the spanish micinn juan de la cierva program and mat2011 - 24331 , the cost action mp1209 and the nsf grant dmr-0844899 .
100 url # 1#1urlprefix[2][]#2 shakouri a 2011 _ annu . rev . mater .
res . _ * 41 * 399431 http://dx.doi.org/10.1146/annurev-matsci-062910-100445 staring a a m , molenkamp l w , alphenaar b w , van houten h , buyk o j a , mabesoone m a a , beenakker c w j and foxon c t 1993 _ europhysics letters ( epl ) _ * 22 * 5762 issn 0295 - 5075 , 1286 - 4854 http://iopscience.iop.org/0295-5075/22/1/011 svensson s f , hoffmann e a , nakpathomkun n , wu p m , xu h q , nilsson h a , snchez d , kashcheyevs v and linke h 2013 _ new j. phys . _ * 15 * 105011 issn 1367 - 2630 http://iopscience.iop.org/1367-2630/15/10/105011 koch j , von oppen f , oreg y and sela e 2004 _ phys .
b _ * 70 * 195107 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/abstract/prb/v70/e195107 kubala b , knig j and pekola j 2008 _ phys .
lett . _ * 100 * 066801 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/abstract/prl/v100/e066801 wirkowicz r , wierzbicki m and barna j 2009 _ phys .
b _ * 80 * 195409 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/doi/10.1103/physrevb.80.195409 snchez r , lpez r , snchez d and bttiker m 2010 _ phys .
_ * 104 * 076801 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/doi/10.1103/physrevlett.104.076801 zhang y , dicarlo l , mcclure d t , yamamoto m , tarucha s , marcus c m , hanson m p and gossard a c 2007 _ phys .
lett . _ * 99 * 036603 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/doi/10.1103/physrevlett.99.036603 braggio a , knig j and fazio r 2006 _ phys .
_ * 96 * 026805 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/abstract/prl/v96/e026805 flindt c , novotn t , braggio a , sassetti m and jauho a p 2008 _ phys .
lett . _ * 100 * 150601 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/abstract/prl/v100/e150601 esposito m , harbola u and mukamel s 2009 _ rev . mod .
phys . _ * 81 * 1665 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/doi/10.1103/revmodphys.81.1665 snchez r , platero g and brandes t 2008 _ phys . rev .
b _ * 78 * 125308 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/doi/10.1103/physrevb.78.125308 van wees b j , kouwenhoven l p , willems e m m , harmans c j p m , mooij j e , van houten h , beenakker c w j , williamson j g and foxon c t 1991 _ phys .
b _ * 43 * 1243112453 http://link.aps.org/doi/10.1103/physrevb.43.12431 koch j , von oppen f and andreev a v 2006 _ phys . rev .
b _ * 74 * 205438 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/abstract/prb/v74/e205438 koch j , semmelhack m , von oppen f and nitzan a 2006 _ phys
b _ * 73 * 155306 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/abstract/prb/v73/e155306 ojanen t and jauho a p 2008 _ phys .
lett . _ * 100 * 155902 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/doi/10.1103/physrevlett.100.155902 ruokola t , ojanen t and jauho a p 2009 _ phys .
b _ * 79 * 144306 copyright ( c ) 2010 the american physical society ; please report any problems to [email protected] http://link.aps.org/doi/10.1103/physrevb.79.144306 ingold g l and nazarov y v 1992 charge tunneling rates in ultrasmall junctions _ single charge tunneling _ ( _ nato asi series b _ vol 294 ) ed grabert h and devoret m h ( new york : plenum press ) pp pp | we review recent theoretical work on thermoelectric energy harvesting in multi - terminal quantum - dot setups .
we first discuss several examples of nanoscale heat engines based on coulomb - coupled conductors .
in particular , we focus on quantum dots in the coulomb - blockade regime , chaotic cavities and resonant tunneling through quantum dots and wells .
we then turn towards quantum - dot heat engines that are driven by bosonic degrees of freedom such as phonons , magnons and microwave photons .
these systems provide interesting connections to spin caloritronics and circuit quantum electrodynamics . |
of all the light that has been emitted by stars , about half has been absorbed by interstellar dust and thermally re - radiated at far - infrared to submillimeter wavelengths , appearing as a diffuse , extragalactic , cosmic infrared background spanning 11000@xmath0 m ( cib ; * ? ? ?
* ; * ? ? ?
statistically characterizing the sources responsible for this background is necessary to gain a full understanding of galaxy formation and cosmology , and thus remains an ongoing pursuit .
the cib was first detected in spectroscopy with the far infrared absolute spectrophotometer ( firas ; * ? ? ?
* ; * ? ? ?
observations of local starburst galaxies with the _ infrared astronomical satellite _ ( _ iras _ ; * ? ? ?
* ) showed that galaxies could emit a surprisingly large part of their energy in the far - infrared , and ground - based measurements later confirmed the existence of a rare population of extremely luminous submillimeter galaxies ( or smgs ; * ? ? ?
* ) . while these bright objects generated tremendous excitement , and in fact constitute a significant fraction of the total star - formation rate density at @xmath11 ( e.g. , * ? ? ?
* ; * ? ? ?
* ) , their low abundance only accounts for a small fraction of the total cib .
the arrival of the _ herschel space observatory _
whose instruments , pacs ( 70 , 100 , and 160@xmath0 m ; * ? ? ?
* ) and spire ( 250 , 350 , and 500@xmath0 m ; * ? ? ?
* ) , bracket the peak of the thermal spectrum of dust emission brought the promise of directly detecting less luminous and far more numerous dusty star - forming galaxies ( dsfgs ) .
however , source confusion resulting from the relatively large point - spread functions ( e.g. , * ? ? ?
* ) limited the number of galaxies that could be individually resolved by pacs at 100 and 160@xmath0 m to 75% @xcite and 74% @xcite of the cib , respectively , and by spire at 250@xmath0 m to 15% @xcite .
statistical methods including stacking ( e.g. , * ? ? ?
* ; * ? ? ?
* ) and _ p(d ) _
@xcite performed better , resolving 70% at 250 , 350 , and 500@xmath0 m for the former , and 89% and 70% of the cib at 100 and 250@xmath0 m for the latter , respectively .
the origin of the rest remained unclear .
@xcite suggested that the missing flux could be tied up in faint sources faint either because they are low mass , at high redshift , or extremely dusty .
another possible source is diffuse emission from the dust that is known to be distributed in the halos of galaxies @xcite , and could be heated by stripped stars ( e.g. , * ? ? ?
* ; * ? ? ?
* ) . here
we present a technique to demonstrate that most , if not all , of the cib can be accounted for by the combined emission from galaxies detected in current near - infrared surveys , and their faint companion objects , at @xmath12 .
we show with a simple model that galaxies alone are the most plausible source of this signal , and argue that any remaining cib likely originates from galaxies at still higher redshifts .
we perform our analysis on catalogs and images located in the cosmos field @xcite , centered at @xmath13 .
we use the @xmath14-selected , publicly - available catalog from ( * ? ? ?
* ultravista ) , which consists of a @xmath15 subset of the full cosmos field , where both near - infrared and optical wavelengths are available ( 30 bands in all ) .
the catalog contains photometric redshifts computed with eazy @xcite , and stellar masses computed with fast @xcite .
galaxies are split into star - forming or quiescent based on their positions in the rest - frame @xmath16 vs. @xmath17 color - color diagram ( _ uvj _ ; * ? ? ?
we use submillimeter maps observed with the spectral and photometric imaging receiver ( spire ; * ? ? ?
* ) at 250 , 350 , and 500@xmath0 m from the herschel multi - tiered extragalactic survey ( hermes ; * ? ? ?
cosmos is a level 5 field , consisting of 4 repeat observations , to an instrumental depth of 15.9 , 13.3 , and 19.1mjy ( 5@xmath18 ) , with confusion adding an additional noise term of 24.0 , 27.5 , and 30.5mjy ( 5@xmath18 ) at 250 , 350 , and 500@xmath0 m , respectively @xcite .
absolute calibration is detailed in @xcite , with calibration uncertainties of @xmath19 .
maps are made with 4arcsec pixels at all wavelengths using the smap @xcite pipeline .
spire maps are chosen specifically for this study because its wavelengths probe the rest - frame peak of thermal dust emission at @xmath203 @xcite , and because large - scale features can be reconstructed with minimal filtering @xcite .
[ fig : tcib ]
we now present a novel method to estimate the extent to which known sources and their faint companions contribute to the cib .
we do this by exploiting an inherent weakness of stacking : that stacking on images with poor angular resolution can result in a boosted ( or biased ) average flux density arising from faint , uncatalogued , companion galaxies ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the trick is in recognizing that only _ correlated _
( i.e. , clustered ) sources will bias the stacked flux density ( for an in - depth discussion see * ? ? ?
* ; * ? ? ?
* ) , so that summed intensities estimated with increasingly smoothed maps places limits on the full intensity in a given redshift range .
meanwhile , emission that is not correlated say , emission coming from sources at redshifts greater than those of our catalog objects will not influence the primary measurement , except as a potential noise term .
for this analysis we use the publicly - available simstack code , which is described in detail in @xcite . briefly , synthetic images of the sky are constructed from _ correlated _ subsets of catalogs ( i.e. , objects in the same redshift range ) with the assumption that galaxies that are physically similar in this case quiescent or star - forming galaxies within a stellar mass bin have comparable infrared luminosities and submillimeter flux densities .
synthetic images are then convolved with the psf of the instrument , and are fit simultaneously to the actual sky map to retrieve the mean flux densities of the subsample . to induce a bias we smooth the maps by convolving them with gaussians whose widths are the geometric differences of the nominal spire and effective beams , i.e. , @xmath21 , where the nominal spire resolutions are 17.5 , 23.7 , and 34.6arcsec full width at half - maximum ( fwhm ) at 250 , 350 , and 500@xmath0 m , respectively , and the effective resolutions are 20 , 30 , 40 , 60 , 80 , 110 , 140 , 180 , 220 , 260 , and 300arcsec fwhm . above 300arcsec , statistical uncertainties become consistent with zero ( see section [ sec : results ] ) .
all synthetic images and sky maps are mean - subtracted before stacking .
following @xcite we split the sample into @xmath22 bins of redshift ( @xmath23 to 4 with @xmath24 ) and stellar mass ( 5 star - forming and 3 quiescent ) , and stack the subsamples with simstack at each smoothing scale .
stacked flux densities are then color - corrected to account for the observed spectral shape of the sources through the passbands , with temperatures taken from ( * ? ? ?
* equation 18 ) .
color corrections range between , 1.0060.984 , 0.9930.977 , and 1.0070.991 at 250 , 350 , and 500@xmath0 m , respectively .
[ fig : stackedflux ] we use simulations to check that stacking and smoothing with a gaussian , as opposed to the measured spire beam or any other kernel , is a reasonable approximation , finding a bias of less than 4% for the largest smoothing kernel , which we include in the reported errors .
we test that the method does not introduce unintended biases by performing 100 null tests for each set of synthetic images and maps .
null tests involve running the identical stacking pipeline with the same binning of sources , but after randomizing the positions of the sources in the catalogs . because the map and images are mean - subtracted , we expect the stacked flux densities of the null tests to be consistent with zero . in section
[ sec : results ] we show that this is the case .
we note that this method has the advantage that missing sources are not double - counted , meaning that the flux density from a single missing object will be distributed among the synthetic images , rather than appearing multiple times ( as would be the case in thumbnail stacking ) .
also note that stacked flux densities are intentionally _ not _ corrected for completeness because it is precisely the flux densities of the missing ( i.e. , incomplete , but similarly applies to misclassified agn or dsfgs ) sources that we are attempting to measure by degrading the effective resolution of the map . as a result , this technique is not limited to cib studies , but is applicable to any study where estimates of the level of faint , correlated emission are in question .
we report total intensities of our stacking measurement as the cumulative sum over redshift ( from @xmath25 to the redshifts labeled with different color lines ) vs. the effective resolution of the image , in figure [ fig : tcib ] . at lower redshifts ( @xmath26 )
the fractional cib that is measured increases weakly with increasing effective beam size , which is expected given that the completeness of the catalog at these redshifts is near unity for stellar masses log(@xmath27 .
conversely , at higher redshifts the fractional cib that is measured increases rapidly with increased smoothing .
the maximum cib we resolve is @xmath2 , @xmath3 , @xmath4 at 250 , 350 , and 500@xmath0 m .
we note slightly different behaviors between the three bands ; particularly at 500@xmath0 m , which appears to have not converged , and may be indicative of higher redshift contributions to the cib .
estimates of the exact measured fraction of the absolute cib are limited by the uncertainties in the reported absolute values of the cib derived from firas spectra by @xcite and @xcite , which range between 22% and 30% .
the @xcite levels ( @xmath28 , @xmath29 , @xmath30 ; hereafter chosen to represent the fiducial values ) are shown as dashed lines in figure [ fig : tcib ] , and the @xcite levels ( @xmath31 , @xmath32 , @xmath33 ) are shown as 3-dot - dashed lines . additionally , ( * ? ? ?
* dot - dashed lines ) provide estimates of the total cib by extrapolating their measured counts , finding they agreed with firas with similarly large uncertainties .
in total we find that the fraction of the @xcite fiducial cib we resolve is @xmath6 , @xmath7 , and @xmath8 at 250 , 350 , and 500@xmath0 m , respectively .
the full set of 100 null tests at each effective beam size are shown as gray dotted lines in figure [ fig : tcib ] , with the 1@xmath18 limits represented by the shaded grey region .
for clarity only the tests for @xmath12 are shown .
as expected , the results of the null tests fluctuate around zero , with the magnitude of that fluctuation increasing with increasing effective beam size .
additional systematic uncertainties include calibration and beam area uncertainties , and cosmic variance , which is estimated following @xcite . , combine parametric descriptions for the stacked flux densities ( at the native resolution of the images ) and the stellar mass functions ( see figure [ fig : stackedflux ] ) .
extrapolated counts from @xcite are shown as gray squares . also shown
are cumulative cib intensities vs. redshift on stacks made with the native ( i.e. , non - smoothed ) spire images ( diamonds ) , and the models after they have been modified to reflect the incompleteness of the actual catalog ( dashed lines ) . ,
scaledwidth=51.0% ] [ fig : didz ]
the cib can be divided into three components : ( 1 ) the contribution from the sources in the parent @xmath34 catalog ; ( 2 ) the contribution from correlated companion sources that for some reason do not make the catalog cut but are recovered with our smoothing - stacking method ; and ( 3 ) uncorrelated emission , including and likely dominated by sources lying at @xmath35 .
it is interesting to consider the nature of ( 2 ) , the undetected sources that are captured by our smoothing procedure : are they very low - mass galaxies ; massive but dusty galaxies that evade detection in current deep @xmath36-selected catalogs ; or some other unknown component ?
we now propose through a simple model that combines a parametric description of the stellar mass function with simple fits to nominal stacked flux densities that the galaxies that make up the stellar mass function are alone able to describe our measurement , and that by extension the missing cib can reasonably be attributed to galaxies in the low - mass end of the stellar mass function .
the first component of the model adopts the ( * ? ? ?
* equation 1 ) parameterization of the @xcite stellar mass function .
the rightmost panel of figure [ fig : stackedflux ] illustrates the performance of the @xcite model ( solid lines ) against the @xcite data ( circles ) .
we find that it diverges from the measurements at higher redshifts and so we add a modification at @xmath37 ( dotted lines ) by simply interpolating between the model and the measured stellar mass functions of @xcite to @xmath38 .
we check that the exact value of faint - end slope at high-@xmath39 negligibly affects the integral , and set it to -1.6 .
similarly , the second component of the model consists of power - law fits to the stacked flux densities vs. lookback time for each stellar mass bin , with the added condition that the mass dependence of the slopes and offsets themselves follow smooth functions .
the three left panels of figure [ fig : stackedflux ] show the stacked flux densities and best - fit power laws as open circles with error bars , and dotted lines , respectively . finally , the model is integrated over the redshift range @xmath404 , and stellar mass range log(@xmath4114 ; although the contribution from galaxies below log(@xmath42 is found to be negligible .
figure [ fig : didz ] compares the resulting model ( solid lines ) with the full cib measurements ( open circles ) at all three wavelengths .
note that model is not a fit , and yet describes the measurement remarkably well , demonstrating that the faint - end of the mass function can plausibly explain the recovered cib .
also in good agreement are the ( * ? ? ?
* gray squares ) estimates derived through extrapolation of their number counts .
we note some tension with the @xcite intensity at 350@xmath0 m ( @xmath43 ) , which is an output of their halo model fit to cib power spectra ; although their 550@xmath0 m estimate from the same model is in relative agreement ( @xmath44 ) . also shown ( as diamonds )
are measurements made when stacking on the native spire images ( i.e. , no smoothing ) .
they are again compared to the model ( dashed lines ) , but this time the model is modified so that the low - mass limit of the integral over the mass function effectively begins at higher masses with increasing redshift , reflecting the completeness behavior of the catalog .
the difference between the solid and dashed lines can be interpreted as a measure of how much of the cib originates from the parent sample , and how much is from companion sources not detected at this @xmath45 limit .
if , for arguments sake , the catalog were 100% complete at all stellar masses and all redshifts , then the dashed and solid lines would overlap .
arguments for additional , diffuse , sources of cib ; in particular dust in the extended halos of galaxies @xcite , are thus disfavored . while our measurements are consistent with the fiducial levels of the total cib , the existing uncertainties on the absolute level are so large that it is difficult to convincingly estimate how much cib is still unresolved .
any missing cib is more likely to occur at longer wavelengths , which are more sensitive to higher redshifts ( because of the negative @xmath36-correction ; * ? ? ?
* ) , from where we would expect uncorrelated emission to originate .
indeed , several exceptional ulirg - like galaxies have been identified at @xmath10 , albeit with low abundances ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
however , extrapolations of the contribution of ulirgs to the star - formation rate density at @xmath10 ( e.g. , * ? ? ?
* ; * ? ? ?
* ) points to them dominating the far - infrared emission at this epoch . determining the relative levels that they and less luminous galaxies contribute is a key question going forward . to resolve these high-@xmath39 questions ,
more data are required ; particularly : 1 ) an update of the absolute cib ; 2 ) deeper catalogs with stellar masses and redshifts to redshifts greater than 4 ; 3 ) submillimeter surveys ( 350 to 1000@xmath0 m ) with large angular - scale fidelity and smaller psfs in the regions of those deep catalogs particularly at longer wavelengths which are more sensitive to galaxies at higher redshifts .
the former can only be achieved with space - based missions to measure the dc level above atmospheric foregrounds .
the second requirement is steadily growing from multiple current or upcoming efforts ( e.g. , sdss / boss , candles , des , lsst ) , although the photometric redshifts for very dusty galaxies may remain quite uncertain ( see * ? ? ?
the last point will require a ground - based submillimeter observatory similar in scope to ccat @xcite .
we find that most of the cib at 250 , 350 , and 500@xmath0 m can be accounted for by galaxies detected in current near - infrared surveys of moderate depth ( @xmath46 ) , and galaxies correlated with them .
we report total intensities of @xmath2 , @xmath3 , @xmath4 at 250 , 350 , and 500@xmath0 m , which corresponds to a fraction of the @xcite absolute cib of @xmath6 , @xmath7 , and @xmath8 .
we find that a simple model combining parametric descriptions for the stellar mass function and stacked flux densities for log(@xmath9 and @xmath12 is able to convincingly reproduce the measurements , which supports the argument that the sources in the faint - end of the mass function make up the previously missing cib .
we note that emission from objects that are not in the catalog , _ and _ is either uncorrelated with catalog objects or exists at scales greater than 300arcsec , would be missed .
however unlikely , this can not be ruled out without a better absolute measurement to compare against .
finally , we propose that any remaining cib likely originates from galaxies at @xmath10 , and if so should be detectable at submillimeter and millimeter wavelengths .
mpv warmly thanks charlotte clarke , pete hurley , seb oliver , and the university of sussex for their hospitality during the development of this study ; and phil hopkins for valuable discussions of the @xmath47 universe .
we also thank the anonymous referee , whose careful comments greatly improved this paper .
spire has been developed by a consortium of institutes led by cardiff univ .
( uk ) and including : univ .
lethbridge ( canada ) ; naoc ( china ) ; cea , lam ( france ) ; ifsi , univ .
padua ( italy ) ; iac ( spain ) ; stockholm observatory ( sweden ) ; imperial college london , ral , ucl - mssl , ukatc , univ .
sussex ( uk ) ; and caltech , jpl , nhsc , univ .
colorado ( usa ) .
this development has been supported by national funding agencies : csa ( canada ) ; naoc ( china ) ; cea , cnes , cnrs ( france ) ; asi ( italy ) ; mcinn ( spain ) ; snsb ( sweden ) ; stfc , uksa ( uk ) ; and nasa ( usa ) .
the spire data for this paper were obtained as a part of proposal kpgt_soliver_1 , with images made using the following obsids : 1342222819 - 26 , 1342222846 - 54 , 1342222879 - 80 , 1342222897 - 901 . , t. a. , giovanelli , r. , radford , s. , & zmuidzinas , j. 2006 , in society of photo - optical instrumentation engineers ( spie ) conference series , vol .
6267 , society of photo - optical instrumentation engineers ( spie ) conference series , 2 | we report contributions to cosmic infrared background ( cib ) intensities originating from known galaxies and their faint companions at submillimeter wavelengths . using the publicly - available ultravista catalog , and maps at 250 , 350 , and 500@xmath0 m from the _ herschel _ multi - tiered extragalactic survey ( hermes )
, we perform a novel measurement that exploits the fact that uncatalogued sources may bias stacked flux densities particularly if the resolution of the image is poor and intentionally smooth the images before stacking and summing intensities . by smoothing the maps we are capturing the contribution of faint ( undetected in @xmath1 ) sources that are physically associated , or _ correlated _ , with the detected sources .
we find that the cumulative cib increases with increased smoothing , reaching @xmath2 , @xmath3 , and @xmath4 at 250 , 350 , and 500@xmath0 m at @xmath5 full width at half - maximum .
this corresponds to a fraction of the fiducial cib of @xmath6 , @xmath7 , and @xmath8 at 250 , 350 , and 500@xmath0 m , where the uncertainties are dominated by those of the absolute cib .
we then propose , with a simple model combining parametric descriptions for stacked flux densities and stellar mass functions , that emission from galaxies with log(@xmath9 can account for the most of the measured total intensities , and argue against contributions from extended , diffuse emission .
finally , we discuss prospects for future survey instruments to improve the estimates of the absolute cib levels , and observe any potentially remaining emission at @xmath10 . |
supersymmetric yang - mills quantum mechanics ( symqm ) provides the quantitative model of m - theory @xcite . even though much simpler than the original theory the model
is not solved in spite of its long history@xcite .
we have therefore decided to set up a systematic lattice survey of symqm beginning with the simplest case of @xmath1(quenched)@xcite and gradually extending it as far as possible towards the bfss limit _
i.e. _ @xmath2 and @xmath3 . in this talk
i will report on the second step along this programme : the first results for higher n will be presented .
the action of the symqm reads @xmath4 where @xmath5 , and all fields are independent of the space coordinates @xmath6 .
the supersymmetric fermionic partners belong to the addjoint representation of su(n ) . the discretized system is put on a @xmath7 dimensional hypercubic lattice @xmath8 which is reduced in all space directions to @xmath9 , @xmath10 .
the gauge part of the action has now the usual form s_g=-_m=1^n_t _
> 1n re ( u_(m ) ) , [ sg ] with = 2n / a^3 g^2 , [ beta ] and @xmath11 , @xmath12 , where @xmath13 denotes the lattice constant and @xmath14 is the gauge coupling in one dimension .
the integer time coordinate along the lattice is @xmath15 .
periodic boundary conditions @xmath16 , @xmath17 , guarantee that wilson plaquettes @xmath18 tend , in the classical continuum limit , to the appropriate components @xmath19 without the space derivatives .
up to date we have addressed the two problems : 1 ) extracting the continuum limit from the lattice data , and 2 ) the search for the nontrivial phase structure .
the first point is essential in any approach based on the discretization . in particular restoration of the continuum supersymmetry , of the full unquenched model ,
may crucially depend on the ability to control the continuum limit .
the second issue is connected to the problem of the bekenstein - hawking entropy which has an elegant solution in the framework of m - theory @xcite .
namely the supersymmetric , extremal black holes found in the latter can be viewed as composed of the elementary d0 brane excitations , providing the statistical interpretation of the area of the `` schwarzschild '' horizon which is known to behave as an etropy . in particular , the theory also predicts existence of the two phases in which the gravity and the elementary d0 branes provide good description respectively@xcite .
we have found in @xcite that the continuum limit of the model can be readily extracted with the bare parameters scaling with canonical dimensions .
this is expected for the one dimensional system . to search for the phase transition we have studied the distribution of the eigenvalues of the polyakov line @xmath20 , which is a very sensitive determinant of the phase structure in gauge theories .
it was found that , similarly to the large volume qcd , in the low temperature phase the eigenvalues are concentrated around zero , while at high temperature the distribution is peaked around @xmath21 which constitute the center of su(2 ) . in the space
extended theories the @xmath22 symmetry is spontaneously broken in the infinite volume limit and only one direction is populated . in the present 0-volume system , this may happen only in the infinite - n limit , the gross - witten model being a known example of the critical behaviour emerging at large n.
for higher groups we find now the same behaviour , see fig.1 . since @xmath23 , the histograms correspond to the low and high temperature regions for a range of @xmath24 .
evidently they change from convex to concave at some critical value of @xmath25 similary to the n=2 case .
the nature of the transition is not resolved yet .
nevertheless , our data show unambiguously that the system behaves differently in both regions .
for example , we have also measured the dependence of the size of the system , @xmath26 , on the temperature , and found that it is definitely different , and the change in the behaviour occurs at the same t where the distributions in fig .
1 change their shapes .
it is also possible , for the first time to confront the n dependence of the transition temperature with theoretical expectations .
it follows from eq.([beta ] ) and @xmath27 , that the t hooft scaling @xmath28 implies that the lattice coupling @xmath29 at fixed @xmath30 .
fig.2 shows the @xmath24 dependence of @xmath31 for available range of n together with the fit of the first @xmath32 correction .
indeed , the reduced critical coupling seems to saturate towards higher n and one can estimate that @xmath33 result is within @xmath34 15% of the @xmath35 one ( a horizontal line ) .
we have also studied the new , noncompact formulation of the model which has better numerical behaviour @xcite . in this approach
the @xmath36 spatial degrees of freedom are noncompact @xmath37 and are defined at the discrete time intervals , while the temporal one remains compact @xmath38 .
the action @xmath39 reads @xmath40 the covariant finite difference along the time direction @xmath41 takes into account the parallel transport between adjacent lattice sites .
this system has the same local gauge invariance as the compact version , eq.([sg ] ) . with the new action
we have extended previous study to higher dimensions , d=6 and d=10 .
preliminary simulations confirm results found for d=4 . in particular
the average size of the system @xmath42 shows a characteristic break in the temperature dependence at the position consistent with the t hooft scaling .
moreover , the @xmath42 decreases with d in agreement with the mean field results @xcite .
quantitative lattice study of the yang - mills quantum mechanics , and possibly the m - theory , have just begun .
quenched results are encouraging , but a lot remains to be done .
simulations work for all interesting values of the dimension @xmath7 and are feasible for a range of n. recent results give us a rough idea how the large n limit is approached and where the asymptotics sets it .
all quenched simulations performed up do date indicate existence of the two regions at finite temperature .
this intriguing correspondence with the predictions of the m - theory should be further quantified .
of course , the next step is to include the dynamical fermions .
this can be done by a brute force for d=4 and for the first few n s at d=10
. the one dimensional nature of the system should help considerably . for higher n , at d=10 ,
we face the problem of the complex pfaffian . an important insight into the whole subject may be gained by applying the full potential of the small volume approach @xcite .
9 t. banks , w. fishler , s. shenker and l. susskind , .
m. claudson and m. b. halpern , .
u. h. danielsson , g. ferretti and b. sundborg , . m. b. halpern and c. schwartz , . r. a. janik and j. wosiek , ; hep - lat/0011031 .
a. strominger and c. vafa , .
e. j. martinec , hep - th/9909049 .
p. bialas and j. wosiek , hep - lat/0109031 .
d. kabat , g. lifszytz and d. a. lowe , .
m. lscher , ; p. van baal , ; p. van baal , hep - ph/0008206 . | we present new results of the quenched simulations of the reduced d=4 supersymmetric yang - mills quantum mechanics for larger gauge groups su(n ) , @xmath0 .
the model , studied at finite temperature , reveals existence of the two distinct regions which may be precursors of a black hole and the elementary d0 branes phases of m - theory conjectured in the literature .
present results for higher groups confirm the picture found already for n=2 .
similar behaviour is observed in the preliminary simulations for the d=6 and d=10 models . |
the karlsruhe tritium neutrino experiment katrin @xcite will be the first beta decay experiment attempting to measure the electron neutrino mass with sub - ev precision .
presently the experiment is commissioned to start data - taking in 2013/14 and has a projected sensitivity of 0.2 ev ( 90% c.l . ) to the neutrino mass .
katrin is the successor of the experiments in mainz @xcite and troitsk @xcite and will be using some of the same techniques as those . for the technical details of katrin
see e.g. @xcite . strictly speaking when measuring the electron neutrino mass with @xmath2-decay spectra ,
what we get is the socalled kinematic neutrino mass .
that is , the incoherent sum of neutrino mass eigenvalues weighted by the appropriate entries in the lepton mixing matrix : @xmath3 however , because the mass differences between the active neutrino mass states are known to be smaller than katrins sensitivity the experiment can effectively only see one mass state ( the mass squared differences are @xmath4 ev@xmath5 and @xmath6 ev@xmath5 , respectively @xcite ) .
this mass state is sometimes called the electron neutrino mass , but in principle the tritium beta - spectrum could contain the signatures of more than one mass state or of couplings to other particles entirely . in order to be called truly model independent katrin s final data should be analyzed also for alternative scenarios - beyond the minimal extension of the standard model . performing
an analysis for non - standard couplings to the electron neutrino adds more parameter space to the @xmath1-function of the experiment .
one should therefore consider how an extended analysis should be performed on the katrin output in order to get reliable results .
we present here one approach which seems to give several advantages over the standard frequentist analysis . in section 2
we describe our analysis methods before presenting results for a number of cases in section 3 .
finally we give some concluding remarks in the last section .
let us begin by summarising the procedures for production and analysis of katrin spectra as performed by a toy model monte carlo and analysis code for katrin - like experiments @xcite .
this code has previously been used to forecast the experiment s sensitivity to the neutrino mass @xcite .
because katrin has an integrating spectrometer ( a consequence of the mac - e ( magnetic adiabatic collimation with electrostatic ) filter technique ) the beta - spectrum must be written as an integral over the electron energy : @xmath7 here @xmath8 is the retarding potential of the spectrometer , @xmath9 is the total number of tritium nuclei in the source , @xmath10 is the measurement time allotted for a given value of the retarding potential and @xmath11 is the experimental response function ( which in turn is a combination of the electron energy loss function of the tritium source and the transmission function of the spectrometer).@xmath12 is the theoretical beta spectrum rate folded with the electronic final state distribution of molecular tritium . a retarding voltage - independent background rate of @xmath13 is now added to eq . ( [ eq : tid ] ) : @xmath14 this gives us the following theoretical expression for a katrin - like spectrum : @xmath15 individual spectra ( to resemble the real measurements )
are built using initial parameters and @xmath16 for the neutrino mass squared endpoint energy , respectively . to the theoretical expression
is then added a random component from a gaussian distribution with @xmath17 and @xmath18 : @xmath19 when we want to fit our randomized beta - spectra we have to account for statistical fluctuations by allowing the overall amplitude @xmath20 of the signal as well as the background rate @xmath21 to vary against the theoretical amplitude @xmath22 and background rate @xmath13 . in addition , we allow the neutrino mass squared as well as the endpoint energy @xmath23 to deviate from the initial parameters of the simulation and @xmath16 .
@xmath24 combining eq.s ( [ eq : best2 ] ) and ( [ eq : best3 ] ) we finally get katrin s @xmath1-function @xcite : @xmath25 beta - spectra[fig : fig1 ] for a katrin - like parameter set and an assumed value for the neutrino mass of @xmath26 measured with an optimized time distribution over the last 25 ev of the beta - spectrum , e.g. compare to @xcite.,width=597 ] the analysis of the simulated data can be performed with minuit2 which is imbedded in the root - package .
this procedure performs a minimization of the @xmath1-function using combinedminimizer .
combinedminimizer in turn uses either an evaluation of the covariance matrix or a simplex method to find the best minimum of the @xmath1-function in the parameter space @xcite .
one can now do a standard frequentist analysis to find the statistical uncertainty on e.g. the neutrino mass by producing a suitable amount of monte carlo spectra , performing the minimization for each of them and finally inspecting the resulting histograms .
an example is shown in figure [ fig : fig1 ] for 12860 spectra produced with @xmath27 0.0 ev .
however as previously indicated the minimization approach has a number of drawbacks .
for one thing it does not give any information on multiple minima , and it is not well suited for finding shallow minima .
furthermore extracting detailed information on correlated parameters is pretty laborious .
still the method works just fine for the four well - known free parameters used in a standard katrin analysis - see table [ tab : org ] . but as one adds more parameters the minimization procedure often becomes problematic and rather slow . as an alternative approach we have considered markov chain monte carlo and bayesian inference techniques as in the publicly available cosmomc analysis package for cosmology .
typical cosmological models contain @xmath28 8 - 12 parameters and cosmomc is well suited for relatively fast analysis of such multiparameter spaces @xcite .
the programme is built for analysis of large cosmological datasets such as cmb data from wmap and supernova surveys but can in principle analyse whatever dataset the user provides - the cosmology can be turned off if it is irrelevant .
cosmomc uses bayesian statistics for the analysis .
when doing socalled bayesian parameter inference one is interested in knowing the posterior probability , @xmath29 - the probability of the * parameters * , @xmath30 , given the data @xmath31 and the model @xmath32 .
the inverse question is for the probability of the * data * , d , given the parameters and the model , @xmath33 - this is simply the likelihood function . with these two probabilities and the well - known bayes theorem ,
@xmath34 one can write an expression for the posterior probability : @xmath35 here @xmath36 is the likelihood - which can be easily derived from the @xmath1-function is connected to the @xmath1-function as in the following way : @xmath37 . ] .
the posterior probability is thus proportional to the likelihood .
meanwhile @xmath38 is the socalled prior probability sometimes referred to as the subjective input - it is what we believe we know from theory before even taking the data into account .
correspondingly this probability has no dependence on the data .
note that we have been using flat priors on all input parameters in this paper .
finally @xmath39 , the evidence , is in effect only a parameter - independent normalization constant@xcite
. when we want to know best - fit values and confidence levels of specific parameters we can then simply integrate over all the remaining ( nuisance ) parameters .
this is called marginalization and the output is called the marginalized probability for the parameter of interest .
in addition to this rather convenient production of parameter probability distributions , from the bayesian inference approach , cosmomc gives us another great advantage by using a markov chain monte carlo ( mcmc ) to probe the parameter space .
this will provide a very thorough and easy to inspect mapping of the parameter space of interest .
the purpose of the mcmc is to probe the whole parameter space in a randomized manner . to achieve this one implements the metropolis hastings algorithm @xcite consisting of three main steps : firstly an initial point ,
@xmath40 is chosen .
secondly a step is proposed in some random direction , after which the new point is evaluated : @xmath41 . here @xmath42 means the iterative point , and @xmath43 is the proposed addition taken from some proposal density .
finally the procedure decides whether or not to take the step .
the point @xmath44 is accepted if the posterior probability is improved .
that is if @xmath45 if the expression above is @xmath46 the step is accepted with some probability @xmath47 ( rejected with probability @xmath48 ) . in this manner
we generate a set of points @xmath49 , also called a markov chain . for the number of points , @xmath50 ,
going to infinity we thus have a representation of the posterior probability
. the decision procedure of the metropolis hastings algorithm allows the chain to wander away from any local minima and thus potentially discover other minima ( to a degree determined by the value of r , with temperature , @xmath51 . ] . on the other hand it also guarantees
that the parameter space near the minima is very well probed .
furthermore one can perform the analysis on a combination of multiple chains - all started at random positions - and get an even better picture of the behavior of the different parameters in the allowed intervals .
to get rid of un - physical effects from the random starting points one normally allows for a burn - in
i.e. the first part of the markov chain is removed . in our case
the burn - in is 50% of the sample size . before running the programme one must carefully choose stepsizes and parameter ranges .
several settings in both the cosmomc programme as such and in the parameter files can be tweaked to fit ones purpose .
unfortunately it is in principle not possible to determine in any absolute terms whether or not a specific chain has converged @xcite , but various convergence diagnostics have been developed .
for instance when analyzing multiple chains a convergence parameter , @xmath52 , defined as the variance of the chain means divided by the mean of the chain variances , can be evaluated .
if @xmath53 is less than some chosen small number ( in our case 0.03 ) this information is interpreted as good convergence .
when cosmomc has generated the chains we need , the data analysis is performed giving us best - fit values and standard deviations for all the parameters .
additionally cosmomc produces a number of useful matlab - files which can be used to produce @xmath54 and @xmath55 plots of the marginalized distributions .
inspecting this graphical output allows us to determine if the chains have really converged , whether there are multiple minima and perhaps most importantly it shows parameter correlations right away .
if we go through all of this for say a single monte carlo generated beta spectrum we get all the nice advantages mentioned above .
but the analysis of that one spectrum would take many hours as compared to minutes or seconds with the minuit2 procedure and we would mostly just have achieved a much slower evaluation of the best fit values for that particular spectrum .
however if we in stead use the theoretical beta - spectrum ( which should represent the average of infinitely many measurements or monte carlo realizations ) as our input data - but with monte carlo generated errorbars - our best fit values and standard deviations from cosmomc should correspond to the results of the frequentist approach of building histograms for a very large ( going to infinity ) number of measurements . to recap
we implemented our @xmath1-function for katrin - like experiments in cosmomc and simply turned off cosmology .
as data - set we have used the theoretical spectra - for any given model - with the monte carlo generated errorbars of the original code .
the results will be discussed in the following section .
as a first test of our methods we have attempted to reproduce the katrin sensitivity .
we thus generated a tritium beta - spectrum using as input so far only four parameters : the electron neutrino mass squared , the endpoint of the beta - spectrum @xmath16 ] , the background count rate @xmath13 and the signal count rate near the beta - spectrum endpoint , @xmath22 - see eq .
( [ eq : best1 ] ) . the input signal count rate can be calculated as a combination of the column density of the source and the magnetic fields and cross sections of the spectrometer and source . in our case the count rate near the endpoint @xmath23 is included in the analysis code via an amplitude factor @xmath22 as in eq .
( [ eq : best3 ] ) . the exact definition and full calculation of this factor @xmath22
is included in appendix a of reference @xcite .
given katrin s experimental settings the amplitude has the value @xmath57 hz .
we would like to remark , that the value of the endpoint energy @xmath23 needs to be treated as a free parameter to produce realistic fits with respect to fitting of .
up to now the @xmath58he - @xmath58h mass difference is known from precision penning trap experiments with 1.2 ev precision @xcite , but already the fits of the experiments at mainz @xcite and troitsk @xcite would have needed a much more precise input value to justify keeping @xmath23 fixed in the fit .
.katrin standard analysis parameters .
please note , that we show and plot the deviation of the value of the endpoint of the beta - spectrum from the theoretically expected value : @xmath59 ev . [ tab : org ] [ cols="<,^,^,>",options="header " , ] as compared to the case without righthanded couplings and the right figure shows the bias on @xmath60 .
this analysis includes the righthanded coupling strength as a free parameter .
the bias on the mass is as large as 80% while the values of @xmath60 is up to five times as large as for the standard case ( barring the parameter range below katrins sensitivity where the uncertainty on the mass parameter migrates into the b - dimension in the mcmc ) .
[ fig : m_r],title="fig:",width=264 ] as compared to the case without righthanded couplings and the right figure shows the bias on @xmath60 .
this analysis includes the righthanded coupling strength as a free parameter .
the bias on the mass is as large as 80% while the values of @xmath60 is up to five times as large as for the standard case ( barring the parameter range below katrins sensitivity where the uncertainty on the mass parameter migrates into the b - dimension in the mcmc ) .
[ fig : m_r],title="fig:",width=264 ] for the full parameter range of table [ tab : right ] and the right figure is an enlarged version of this plot for @xmath61 ev . in the left figure
we see that we get the output values wrong by more than a factor 10 !
this is exacerbated at mass values just above the katrins sensitivity once again demonstrating how the uncertainty on these two ill - determined parameters is redistributed in the parameter space of the markov chain .
the right figure shows us that when we look beyond the much larger errorbars around @xmath62 ev the output values still fluctuate with errors of order @xmath63 .
[ fig : b_r],title="fig:",width=264 ] for the full parameter range of table [ tab : right ] and the right figure is an enlarged version of this plot for @xmath61 ev . in the left figure
we see that we get the output values wrong by more than a factor 10 !
this is exacerbated at mass values just above the katrins sensitivity once again demonstrating how the uncertainty on these two ill - determined parameters is redistributed in the parameter space of the markov chain .
the right figure shows us that when we look beyond the much larger errorbars around @xmath62 ev the output values still fluctuate with errors of order @xmath63 .
[ fig : b_r],title="fig:",width=264 ] the results shows us firstly that the output mass values fluctuate rather wildly - and in some cases deviate my as much as @xmath64 80 % from the input values as shown in the left panel of figure [ fig : m_r ] . and secondly the statistical uncertainty is up to 5 times larger than in the standard case except in regions where @xmath65 <
0.2 ev as expected from the discussion above . turning to at the output values of the righthanded coupling strength in figure [ fig : b_r ] we get appallingly bad results especially in the @xmath65 = 0.2 ev -region .
from the left hand picture of figure [ fig : b_r ] one might get the impression that the output value of b is returned rather nicely for the larger masses .
however as shown in the right panel of figure [ fig : b_r ] the relative error is still up to @xmath64 60% in some regions . in conclusion
we see from these numerical artifacts that it is extremely difficult to get a good determination of _ both _ the mass and the coupling strength . at least when using fairly large parameter intervals .
given stronger limits the situation would no doubt change . but judging from our cosmomc contours the values in some cases will be pressed to the largest allowed parameter values even when the intervals are as broad as here . in other words - tighter parameter values in this case
merely amounts to a manual setting of the allowed size of the statistical uncertainties .
next we perform the analysis on the same spectra without including the righthanded coupling strength to get an idea of the bias imposed on the neutrino mass in the presence of unaccounted - for righthanded currents .
we present our results in figure [ fig : nor ] but here the analysis has been performed ( on the same spectra ) without the inclusion of the righthanded coupling strength .
clearly the errors on @xmath65 are much better and for realistic @xmath66-values certainly within acceptable @xmath67-ranges .
however we note that @xmath68 is @xmath69 better in the high - b , high - m corners of the right - side plot .
this coincides with a turnover of the bias on the mass in the left - side plot .
figure [ fig : eor ] shows that this behavior takes place because the @xmath70 parameter is being pushed to the maximally allowed values , which should be avoided .
that is the uncertainty on the mass due to the presence of @xmath66 is migrating into the third correlated parameter - the endpoint of the tritium beta spectrum .
[ fig : nor],title="fig:",width=264 ] but here the analysis has been performed ( on the same spectra ) without the inclusion of the righthanded coupling strength .
clearly the errors on @xmath65 are much better and for realistic @xmath66-values certainly within acceptable @xmath67-ranges .
however we note that @xmath68 is @xmath69 better in the high - b , high - m corners of the right - side plot .
this coincides with a turnover of the bias on the mass in the left - side plot .
figure [ fig : eor ] shows that this behavior takes place because the @xmath70 parameter is being pushed to the maximally allowed values , which should be avoided .
that is the uncertainty on the mass due to the presence of @xmath66 is migrating into the third correlated parameter - the endpoint of the tritium beta spectrum .
[ fig : nor],title="fig:",width=264 ] as it turns out we get much better results when we remove the @xmath66-dimension from our cosmomc setup - this time the bias on the mass is no larger than around 12 % .
we notice however that the statistical error drops steeply for high masses and coupling strengths .
inspecting the original cosmomc likelihood contours we see that this is because @xmath70 has been pushed to the edge of the input interval as shown in figure [ fig : eor ] .
this also explains why the bias flips in the same parameter - range instead of becoming monotonically larger for maximal coupling strengths .
the propagation of the uncertainty on @xmath66 into the @xmath70-dimension is straightforward from the already discussed correlations between the @xmath71 and @xmath70 - parameters .
hopefully the upcoming much more precise @xmath58h@xmath58he mass measurements @xcite will be helpful in resolving this issue for the katrin experiment . vs. @xmath72 for the mass range @xmath73 ev to @xmath74 ev ( again going from the upper left corner to the lower right corner ) when the analysis is performed without the inclusion of @xmath66 .
the figure on the left used spectra that was produced with @xmath75 while the figure on the right is for @xmath76 .
the expected output for @xmath70 is zero , but it is clear to see that in this case the @xmath77 -correlation pushes the uncertainty induced in the mass parameter by the physical presence of @xmath66 into the @xmath70 -parameter instead .
[ fig : eor],title="fig:",width=264 ] vs. @xmath72 for the mass range @xmath73 ev to @xmath74 ev ( again going from the upper left corner to the lower right corner ) when the analysis is performed without the inclusion of @xmath66 .
the figure on the left used spectra that was produced with @xmath75 while the figure on the right is for @xmath76 .
the expected output for @xmath70 is zero , but it is clear to see that in this case the @xmath77 -correlation pushes the uncertainty induced in the mass parameter by the physical presence of @xmath66 into the @xmath70 -parameter instead . [
fig : eor],title="fig:",width=264 ] in conclusion the bias induced on the neutrino mass is now within acceptable bounds and agree well with the results found by bonn _
et al _ @xcite .
finally it should be noted that an experiment such as katrin can clearly not be used to put bounds on the size of the righthanded coupling strength at this point .
a precise knowledge of the neutrino mass and the tritium beta - spectrum endpoint @xmath23 would be have to be presupposed before measurements of the tritium beta spectrum could be used to determine @xmath66 .
our attempt at an analysis of simulated katrin data with various additional parameters has shown the following : for the standard case of analysis with regard to one neutrino mass , the mcmc approach is certainly well suited and gives robust results .
the method is very practical when performing analysis for non - standard cases because the cosmomc output lets us inspect the behavior of the parameters and their relation to one another in a straightforward manner .
we have used the method to build a sensitivty - plot for a katrin - like experiment , clearly demonstrating the dominating dependence of the sensitivity on the signal countrate .
further we have learned that for a suitable mass - squared difference an experiment such as katrin should be able to detect the existence of other neutrino mass states . and
finally we have re - evaluated the influence of couplings to righthanded currents in the tritium beta decay and found that ignoring this would maximally induce an error on the neutrino mass of order 10% .
in conclusion we find that our bayesian approach to the analysis of the katrin experiment is certainly competitive to a frequentist approach and that it has several advantages over it when using an already well - developed framework such as cosmomc .
we acknowledge the use of computing resources from the danish center for scientific computing ( dcsc ) and the grant of bmbf under contract 05a08pm1 .
99 a. osipowicz _ et al . _ [ katrin collaboration ] , + `` katrin : a next generation tritium beta decay experiment with sub - ev sensitivity for the electron neutrino mass , '' + arxiv:0109033 [ hep - ex ] .
j. angrik _ et al . _
[ katrin collaboration ] + `` katrin design report 2004 , '' + wissenschaftliche berichte fz karlsruhe 7090 , http://bibliothek.fzk.de/zb/berichte/fzka7090.pdf ch .
kraus , b. bornschein , l. bornschein , j. bonn , b. flatt , a. kovalik , b. ostrick , e. w. otten , j. p. schall , th .
thmmler and ch .
weinheimer , + `` final results from phase ii of the mainz neutrino mass search in tritium @xmath2 decay , '' + eur .
j * c 40 * , 447 - 468 ( 2005 ) + arxiv:0412056v2 [ hep - ex ] .
s. s. masood , s. nasri , j. schechter , m. a. trtola , j. w. f. valle and c. weinheimer , + `` exact relativistic beta decay endpoint spectrum , '' + phys .
c * 76 * 045501 ( 2007 ) + arxiv:0706.0897v1[hep - ph ] .
n. christensen , r. meyer , l. knox , and b. luey , + `` bayesian methods for cosmological parameter estimation from cosmic microwave background measurements , '' + arxiv:0103134 [ astro - ph ] .
a. sejersen riis , s. hannestad , + `` detecting sterile neutrinos with katrin like experiments , '' + jcap02(2011)011 ( 2011 ) .
+ arxiv:1008.1495v2 [ astro - ph ] .
j. bonn , k. eitel , f. gluck , d. sevilla - sanchez and n. titov , + `` the katrin sensitivity to the neutrino mass and to right - handed currents in beta decay , '' + arxiv:0704.3930 [ hep - ph ] .
n. severijns , m. beck and o. naviliat - cuncic , + `` tests of the standard electroweak model in beta decay , '' + rev .
* 78 * ( 2006 ) 991 [ arxiv : nucl - ex/0605029 ] .
j. hamann , s. hannestad , g. g. raffelt , i. tamborra and y. y. y. wong , + `` cosmology seeking friendship with sterile neutrinos , '' + phys . rev . lett . * 105 * , 181301 ( 2010 )
. + arxiv:1006.5276 .
j. hamann , s. hannestad , g. g. raffelt and y. y. y. wong , + `` observational bounds on the cosmic radiation density , '' + jcap * 0708 * , 021 ( 2007 ) .
+ arxiv:0705.0440 [ astro - ph ] .
j. hamann , s. hannestad , j. lesgourgues , c. rampf and y. y. y. wong , + `` cosmological parameters from large scale structure - geometric versus shape information , '' + arxiv:1003.3999 . m. c. gonzalez - garcia , m. maltoni and j. salvado , + `` robust cosmological bounds on neutrinos and their combination with oscillation results , '' + arxiv:1006.3795 .
m. maltoni & t. schwetz , + `` sterile neutrino oscillations after first miniboone results , '' + phys .
d * 76 * , 093005 ( 2007 ) .
s. goswami and w. rodejohann + `` miniboone results and neutrino schemes with two sterile neutrinos : possible mass orderings and observables related to neutrino masses , '' + arxiv:0706.1462v2 [ hep - ph ] .
g. j. stephenson , j. t. goldman and b. h. j. mckellar , + `` tritium beta decay , neutrino mass matrices and interactions beyond the standard model , '' + phys .
d * 62 * , 093013 ( 2000 ) + arxiv:0006095 [ hep - ph ] . | the katrin ( karlsruhe tritium neutrino ) experiment will be analyzing the tritium beta - spectrum to determine the mass of the neutrino with a sensitivity of @xmath0 ev ( 90% c.l . ) .
this approach to a measurement of the absolute value of the neutrino mass relies only on the principle of energy conservation and can in some sense be called model - independent as compared to cosmology and neutrino - less double beta decay . however by model
independent we only mean in case of the minimal extension of the standard model .
one should therefore also analyse the data for non - standard couplings to e.g. righthanded or sterile neutrinos . as an alternative to the frequentist minimization methods used in the analysis of the earlier experiments in mainz and troitsk we have been investigating markov chain monte carlo ( mcmc ) methods which are very well suited for probing multi - parameter spaces . we found that implementing the katrin @xmath1- function in the cosmomc package - an mcmc code using bayesian parameter inference - solved the task at hand very nicely . |
using particles in a manner similar to surfactants in order to stabilize emulsions is very attractive in particular for the food- , cosmetics- , and medical industry to stabilize , e.g. barbecue sauces and sun cremes or in order to produce sophisticated ways to deliver drugs at the right position in the human body .
the microscopic processes leading to the commercial interest can be understood by assuming an oil - water mixture . without any additives ,
phase separation would take place and the oil would float on top of the water . adding small particles , however , causes these particles to diffuse to the interface which is being stabilized due to a reduced surface energy . if for example individual droplets of one phase are covered by particles , such systems are also referred to as `` pickering emulsions '' and are known since the beginning of the last century @xcite .
particularly interesting properties of such emulsions are the blocking of ostwald ripening and the rheological properties due to irreversible particle adsorption at interfaces as well as interface bridging due to particle monolayers @xcite .
recently , there has been a growing interest in particles suspended in multiphase or multicomponent flows @xcite , which led to the discovery of a new material type , the `` bicontinous interfacially jammed emulsion gel '' ( bijel ) @xcite .
the existence of the bijel was predicted in 2005 by stratford et al .
@xcite and experimentally confirmed by herzig et al . in 2007 @xcite .
in contrast to pickering emulsions which consist of unconnected particle stabilised droplets distributed in a second continuous fluid phase , the bijel shows an interface between two continuous fluid phases which is covered by particles .
since the particles used for stabilization have a larger size than surfactant molecules and do not present any amphiphilic properties , concepts developed for the description of surfactant stabilized systems are often not applicable . instead
, theoretical models have to be developed and experiments have to be performed which consider the specific properties of particle - stabilized systems . these include the particle s contact angle , the strong interparticle capillary forces , or the ph value and electrolyte concentration of the solvents @xcite . however , even today our quantitative understanding of solid stabilized emulsions is still far from satisfactory .
computer simulations are promising to understand the dynamic properties of particle stabilized multiphase flows .
however , the shortcomings of traditional simulation methods quickly become obvious : a suitable simulation algorithm is not only required to deal with simple fluid dynamics but has to be able to simulate several fluid species while also considering the motion of the particles and the fluid - particle interactions .
some recent approaches trying to solve these problems utilize the lattice boltzmann method for the description of the solvents @xcite .
the lattice boltzmann method can be seen as an alternative to conventional navier stokes solvers and is well established in the literature .
it is attractive for the current application since a number of multiphase and multicomponent models exist which are comparably straightforward to implement @xcite .
in addition , the method has been combined with a molecular dynamics algorithm to simulate arbitrarily shaped particles in flow and is commonly used to study the behavior of particle - laden single phase flows @xcite .
a few groups combined multiphase lattice boltzmann solvers with the known algorithms for suspended particles @xcite . in this paper
we follow an alternative approach : we present a method based on the multicomponent lattice boltzmann model of shan and chen @xcite which allows the simulation of multiple fluid components with surface tension .
our model generally allows arbitrary movements and rotations of arbitrarily shaped hard shell particles .
it does not require fluid - filled particles and thus does not suffer from unphysical behavior caused by oscillations of the inner fluid @xcite .
further , it allows an arbitrary choice of the particle wettability one of the most important parameters for the dynamics of multiphase suspensions @xcite .
the remainder of the paper is organised as follows : after a description of the shan - chen approach for multicomponent lattice boltzmann simulations and an extension of the lattice boltzmann method to simulate suspensions , a way to combine the two methods is proposed .
the influence of the parameters of the model on the contact angle as a measure of wettability is studied in the following section .
then the suitability of the new method is tested by performing a detailed study of the formation of bijels and pickering emulsions .
the dynamics of the fluid solvents is simulated by a multicomponent lattice boltzmann model following the approach of shan and chen @xcite . here , each component follows a lattice boltzmann equation @xmath0 where @xmath1 is the single particle distribution function for component @xmath2 in the direction @xmath3 ( @xmath4 ) at a discrete lattice position @xmath5 and at timestep @xmath6 . in this work
we use exclusively the so - called d3q19 implementation , where @xmath7 velocities are used on a three dimensional lattice . for simplicity ,
the length of a timestep and the lattice constant are set to 1 , i.e. all units are given in lattice units if not stated otherwise .
@xmath8 is the bhatnagar - gross - krook ( bgk ) collision operator @xcite , @xmath9 which is a relaxation towards the local equilibrium distribution function @xmath10\ ! \ ] ] on a time scale given by the relaxation time @xmath11
@xcite . here
, @xmath12 is the fluid density with reference density @xmath13 and @xmath14 is the macroscopic bulk velocity of the fluid , given by @xmath15 .
@xmath16 are the coefficients resulting from the velocity space discretization and @xmath17 is the speed of sound , both of which are determined by the choice of the lattice .
the kinematic viscosity of the fluid is given by @xmath18 .
the interaction between fluid components @xmath2 and @xmath19 is introduced as a self - consistently generated mean field force @xmath20 where @xmath21 are the nearest neighbors and @xmath22 is the so - called effective mass , which can have a general form for modeling various types of fluids .
we choose @xmath23 @xmath24 is a force coupling constant whose magnitude controls the strength of the interaction between components @xmath2 , @xmath19 and is set positive to mimic repulsion .
the dynamical effect of the force is realized in the bgk collision operator by adding to the velocity @xmath25 in the equilibrium distribution the increment @xmath26 the force also enters the calculation of the actual macroscopic bulk velocity as @xcite @xmath27 in this paper two fluids with identical properties are used which are called `` blue '' ( `` b '' ) and `` red '' ( `` r '' ) . to simplify statements about the fluid ratio at a certain position an order parameter @xmath28 is introduced .
the shan - chen model is a diffuse interface method , where interfaces between different fluids are about four lattice sites wide . for the analysis below we define the interface position to be located where the order parameter vanishes .
pioneering work on the development of an extension to the lattice boltzmann method to incorporate particles has been done by ladd et al .
the method has been applied to suspensions of spherical and non - spherical particles by various authors @xcite .
recently , the inclusion of brownian motion was revisited and clarified in more detail @xcite .
the suspended particles are assumed to be homogeneous spheres with radius @xmath29 . in our implementation of the method
newton s equations for the momentum @xmath30 and the angular momentum @xmath31 are solved with a leap frog integrator to simulate their behavior . here
, @xmath32 is the force acting on a particle , @xmath33 is its mass and @xmath34 its velocity .
@xmath35 is the torque , @xmath36 the moment of inertia and @xmath37 the angular velocity .
the particles are discretized on the lattice and interactions between the fluid and the particles are introduced by marking all lattice sites that are inside the particle as solid nodes for the fluid , at which bounce back boundary conditions are applied @xcite .
bounce back boundary conditions reflect the incoming distributions at a site back to where they came from , so that the streaming step is modified to @xmath38 for all @xmath39 where @xmath40 is not a solid node to @xmath41 for all @xmath39 where @xmath40 is a solid node . here
, @xmath42 is defined as the index corresponding to @xmath43 .
this results in a no - slip boundary located halfway between the fluid and the solid node .
the change of momentum of the fluid that is reflected at the boundary has to be compensated by a momentum change of the particle itself as given by @xmath44 as we assume the length of a time step @xmath45 to be @xmath46 this corresponds to a force @xmath47 and a torque @xmath48 on the particle . here ,
@xmath49 is the vector pointing from the center of the particle to the site of the reflection .
since the particles are not stationary but move over the lattice , the bounce back rule does not correctly reproduce the velocity of the reflected fluid .
it is therefore corrected as @xmath50 where @xmath51 is the velocity of the particle surface on which the fluid is reflected .
this effect also leads to a change in transfered momentum , so that the force acting on the particle is @xmath52 when the particle moves over the lattice , individual lattice sites can be either occupied by it in front of the particle or be released at its back . in case of newly occupied sites ,
the fluid on the site is deleted and its momentum transfered to the particle by adding @xmath53 in case of the particle vacating a lattice site new fluid is created with the initial fluid density and the velocity of the particle surface @xmath51 at the corresponding site : @xmath54 to satisfy conservation of momentum this again leads to a force on the particle , @xmath55 interactions between particles can be taken into account similar to standard molecular dynamics implementations . in the current paper , we only consider hertzian contact forces to mimic hard spheres and a lubrication correction to correct for the limitations of the lattice boltzmann method to describe the hydrodynamics properly on scales below the lattice resolution . when particles collide the resulting forces
are derived from the hertzian potential @xcite @xmath56 with @xmath57 being a constant @xcite .
when two particles move towards each other the lubrication interaction between them results in a force separating the particles .
if there is not at least one lattice site between the particles to resolve the flow , this force is not properly reproduced by the simulation and a lubrication correction has to be added as given by @xmath58 \left ( \frac{1}{\left|\mathbf{r}\right| - 2r_\textnormal{par } } - 1 \right),\ ] ] where @xmath59 is the vector connecting the particle centers and @xmath60 is their respective velocity @xcite .
it is introduced at particle distances smaller than @xmath61 lattice units and limited to it s value at a distance of @xmath62 of a lattice unit to avoid numerical instabilities due to the divergence at @xmath63 .
in order to develop a simulation algorithm for particles in multicomponent flows the previously described methods can be combined as described in the current section . first , when extending the coupling between particles and fluid to multiple components the treatment of lattice sites that are uncovered by the moving particle has to be adapted .
in the original algorithm for a single fluid , such sites are filled with the initial fluid density @xmath64 .
however , re - initializing such sites with multiple fluid components can lead to artefacts since it is not a priori the case that the correct fluid composition should correspond to the initial state of the simulation . for example
, one kind of fluid could appear in a region where only the other fluid is present . to prevent such artefacts we use the average surrounding fluid density @xmath65 where @xmath66 are all indices @xmath39 for which @xmath67 is a non - particle site .
@xmath68 is the number of these sites .
similar to the method described in @xcite , the uncovered sites are re - initialized with @xmath69 following a similar approach as in eqs .
[ eq : newfi ] and [ eq : fcorrsurf ] , i.e. @xmath70 and @xmath71 the second modification of the original algorithms is required to correctly take into account the effect of the fluid - fluid interaction forces on the fluid in the direct vicinity of a particle .
for the calculation of the forces between different fluid components , also the empty lattice sites inside a particle are considered if one follows eq .
[ eqn_shan - chen - f ] . since there are no shan - chen forces acting in the direction from the particle to the fluid , the fluid forms a layer of increased density around the particle .
to avoid this artefact , the outermost layer of lattice sites inside the particle is not kept empty , but is filled with a virtual fluid density which is equivalent to the average of the surrounding densities @xmath72 : @xmath73 this virtual fluid inside the particles does not follow the lattice boltzmann equation , i.e. the advection and collision steps are not applied . further
, the shan - chen like force acting from the fluid surrounding a particle on the particle itself has to be accounted for .
this is implemented by summing up all shan - chen forces @xmath74 acting on every lattice site @xmath75 inside the particle .
this force and the corresponding torque are then added to the particle within the molecular dynamics algorithm .
the forces on the fluid outside the particles are calculated as before with the virtual fluid being treated like a regular fluid in the shan - chen force computation .
this leads to a balanced force on the fluid sites near the particle surface and therefore prevents the formation of a layer of increased density .
this is demonstrated in fig .
[ fig_methode - totaldensity ] , where the left subfigure shows a particle with @xmath76 being filled with virtual fluid while in the right subfigure the particle is not filled with a virtual fluid .
the particle is set at an interface created by two lamellae of red and blue fluids at the center of the shown area .
periodic boundary conditions are applied causing a second interface to appear at the left and right borders of the sketches .
as we use a diffuse interface method for the fluids , the interfaces cover about four lattice sites depicted by the varying grey scale . without the virtual
fluid , the halo of increased density can clearly be seen , while adding the virtual fluid successfully allows to correct for this inconsistency . after 2000 timesteps in the presence of a particle with @xmath76 at an interface at the center of the shown area .
the cut on the left ( a ) shows a particle being filled with virtual fluid , while on the right ( b ) the particle is empty as in the original algorithm . without the virtual fluid ,
the halo of increased density can clearly be seen , while adding the virtual fluid successfully allows to correct for this inconsistency .
the parameters of the simulation were @xmath77 and @xmath78 for both species , the system of size @xmath79 lattice sites was initially divided into two lamellae of width 24 lattice sites with density @xmath80 , respectively .
all units are given in lattice units throughout this paper . ]
the advantage of a virtual fluid inside the particles is that it can be utilized to modify the wettability of the particles . here
, we follow an approach which has been introduced to model hydrophobic fluid - surface interactions for studying flow in hydrophobic microchannels or droplets on surfaces with arbitrary contact angles @xcite .
our approach is equivalent to the method presented in @xcite .
the shan - chen interaction between the particles and the fluids can be modified by tuning the density of the local virtual fluids .
increasing one of them by an amount @xmath81 causes the particle surface to `` prefer ''
this fluid with respect to the other one , i.e. the repulsion between the increased component and the unmodified one increases .
@xmath82 is called `` particle color '' and a positive particle color is defined as an addition of `` red '' fluid , i.e. @xmath83 whereas a negative color corresponds to `` blue '' fluid being added , i.e. @xmath84 in the next section we demonstrate that the particle color can be used to tune the contact angle of the particle surface at an interface in order to resemble specific fluids and solid materials . as an alternative to the virtual fluid , a modified version of eq .
[ eqn_shan - chen - f ] that takes solid lattice sites and fluid - surface interactions into account could be developed , but the approach presented here is simpler to implement and does not have any relevant impact on the performance of the code .
the changing discretization of the particle together with the fluid - surface interaction force leads to slight mass errors during the step in which vacated lattice nodes are refilled with fluid .
this effect is especially strong for small particles and high forces ( large values for @xmath85 ) .
typical test cases have shown that the total mass after very long simulation times ( @xmath86 timesteps ) increases by about one percent .
this can be explained by the simple interpolation for the amount of newly created fluid which is necessary since no analytical solution for the multicomponent shan - chen model is known that describes the density profile at interfaces .
even though the effect is very small , it can be suppressed if newly created fluid densities @xmath87 are scaled with a correction factor which depends on the total mass error @xmath88 up to the current timestep and the total number of lattice sites in the system @xmath89 .
this leads to a modification of eq .
[ eqn_rho_neu ] : @xmath90 the rate of the adaptive correction can be tuned with the parameter @xmath91 . due to the very small mass error
, the correction can act very slowly , but should not be chosen too fast in order to avoid hysteresis effects .
further , eq . [ masscorr ] reduces unphysical density gradients at particle surfaces and thus contributes to the stability of the algorithm . repeating the same test case as above with a correction factor of @xmath92 results in a deviation of the mass of 0.03 percent after @xmath86 timesteps .
the contact angle @xmath93 is a common measure for the wettability of the particle by the two fluid components .
the influence of various simulation parameters on the contact angle is investigated in the current section .
definition of the contact angle @xmath93 for a particle of radius @xmath29 at the interface between the two fluid components .
@xmath94 is the distance from the particle center to the interface . ] in order to measure the contact angle the following setup is used in the simulations : in @xmath95 direction , one half of the lattice is filled with one fluid component , the other half with the second one .
the particle is placed at the interface at @xmath96 and the simulation is started .
the interface between the two components is tracked via the ( linearly interpolated ) position at which the order parameter is zero .
then , the contact angle @xmath93 can be calculated by @xmath97 where @xmath94 is the difference between the interface position and the particle center in the direction perpendicular to the interface ( cf .
figure [ fig_contactangle_def ] ) .
contact angle versus time for a 48x48x256 system with particle color 0.02 , @xmath98 , @xmath99 , and fluid density @xmath100 . after a rapid change from about @xmath101 to about @xmath102
the contact angle stays at a constant value and only shows small oscillations with an amplitude of about @xmath103 . ] to study the time - evolution of the contact angle a system of size 48x48x256 lattice sites is used .
the particle has a radius of 10 lattice sites and a color of 0.02 .
the initial fluid density is 0.6 and @xmath99 .
figure [ fig_contactangle_t ] shows the time dependence of the contact angle . after a short , fast movement at the beginning of the simulation
the contact angle oscillates slightly around a fixed value . here and
for all further graphs in this section , we average the contact angle over the timesteps from @xmath104 to @xmath105 leading to a final value of @xmath106 .
the error is given by the standard deviation of the data .
relating the variation of the angle to a change of the position of the interface on the lattice with regards to the particle center results in @xmath107 lattice units .
the error in the position measurement is very small with respect to the lattice resolution .
figure [ fig_kw_r]a shows the resulting contact angle for different particle sizes between @xmath108 and @xmath109 . for small particle radii the error increases substantially and the measured angle is not equivalent to the one measured for larger particles , but is up to @xmath110 larger . for example , for @xmath108
the contact angle is @xmath111 . for particles larger than @xmath112
the error stays below @xmath113 and the measured angles are in the range of @xmath114 to @xmath115 .
smaller particles are more susceptible to small forces because of their smaller mass , also one has to keep in mind that our lattice boltzmann multicomponent model is a diffuse interface method . since the interface is about four lattice sites wide , small particles are completely inside the interface region . for particles with a non - integer radius
the error and the angle are larger than for particles with integer radii .
this can be adhered to being a discretisation effect as the particles move on the lattice . , and @xmath100 .
the average measured angle and errors given by the standard deviation over timesteps from @xmath116 to @xmath105 are shown .
the contact angle for @xmath108 is @xmath117 , for particles larger than @xmath112 the error stays below @xmath113 and the angles are in the range of @xmath118 to @xmath115
. particles with a non - integer radius show larger contact angles , which can be adhered to a discretization effect .
[ fig_kw_r ] + b ) contact angle versus particle color for @xmath119 .
the data can be fitted with the equation @xmath120 ( dashed line).,title="fig : " ] , and @xmath100 . the average measured angle and errors given by the standard deviation over timesteps from @xmath116 to @xmath105 are shown .
the contact angle for @xmath108 is @xmath117 , for particles larger than @xmath112 the error stays below @xmath113 and the angles are in the range of @xmath118 to @xmath115 .
particles with a non - integer radius show larger contact angles , which can be adhered to a discretization effect .
[ fig_kw_r ] + b ) contact angle versus particle color for @xmath119 .
the data can be fitted with the equation @xmath120 ( dashed line).,title="fig : " ] the dependency of the contact angle @xmath93 on the particle color @xmath82 is shown in fig .
[ fig_kw_f]b .
one can see an almost linear relation between the contact angle and the particle color in the range from a color of @xmath121 ( contact angle @xmath122 ) to @xmath123 ( contact angle @xmath124 ) .
included in the figure is a linear fit given by @xmath120 to stress this linear behavior . the simulations with a particle color of @xmath125 and @xmath126
result in a detachment of the particle from the interface .
thus , it is possible to choose a specific particle color to obtain the related contact angle or to force detachment from one of the fluids . as a next step
we investigate the influence of the strength of the fluid - fluid interaction force on the contact angle . for strong forces determined by large @xmath127
the interface is well defined and the surface tension high .
low @xmath127 cause a low surface tension and thus a more diffuse interface .
when the coupling constant @xmath127 is varied , the contact angle @xmath93 changes as shown in fig .
[ fig_kw_g_br]a .
the stronger the force the stronger the particle is kept at the interface .
if the force is too weak the particle can not be held at the interface anymore . for @xmath128
almost no change to the contact angle can be observed and @xmath93 converges to @xmath129 with an error smaller than @xmath130 . for @xmath131
the contact angle increases dramatically until the particle does not stay attached to the interface at @xmath132 . on the one hand , a well defined contact angle and
well defined interfaces are preferrable requiring large values of @xmath127 . on the other hand
, too large values can cause very high local flow velocities and the lattice boltzmann method can become unstable .
thus , @xmath127 should be chosen as small as possible . .
for @xmath133 a contact angle of @xmath129 with an error smaller than @xmath130 is measured .
the contact angle increases from this value at @xmath134 to @xmath135 at @xmath99 with an error below one degree .
@xmath136 results in a contact angle of @xmath137 and for smaller @xmath127 the particle detaches from the interface .
+ [ fig_kw_g_br ] b ) contact angle versus the initial fluid density @xmath138 .
@xmath139 results in a contact angle of @xmath140 .
reducing @xmath138 causes the contact angle to decrease to @xmath141 at @xmath142.,title="fig : " ] . for @xmath133 a contact angle of @xmath129 with an error smaller than @xmath130
is measured .
the contact angle increases from this value at @xmath134 to @xmath135 at @xmath99 with an error below one degree .
@xmath136 results in a contact angle of @xmath137 and for smaller @xmath127 the particle detaches from the interface .
+ [ fig_kw_g_br ] b ) contact angle versus the initial fluid density @xmath138 .
@xmath139 results in a contact angle of @xmath140 .
reducing @xmath138 causes the contact angle to decrease to @xmath141 at @xmath142.,title="fig : " ] as can be observed in fig .
[ fig_kw_g_br]b , a variation of the initial fluid density @xmath138 has a similar effect as a modification of the coupling constant on the contact angle .
however , @xmath93 only changes from @xmath143 ( @xmath142 ) to @xmath140 ( @xmath139 ) , i.e. the effect is much weaker .
the reason for the lower impact on @xmath93 is given by our particular choice of the effective mass @xmath22 ( see eq . [ eqn_ref_dens ] ) which causes a damping of the interactions for large densities .
the knowledge of the contact angle and particle shape together with a measurement of the surface tension between both fluids allows to measure the energy required to detach a trapped particle from an interface @xcite . while for spherical particles it is straightforward to compute the detachment energy analytically , for highly anisotropic or complex shaped particles this is not easily possible . a simulation study based on
the model proposed in this paper would allow a well founded understanding of the dependence of detachment energies on particle properties and could be compared to experimental data .
the formation of a `` bijel '' ( bicontinuous interfacially jammed emulsion gel ) was first predicted by stratford et .
al . in 2005
@xcite . as stated in the introduction
, bijels can form when ( colloidal ) particles are added to a mixture of two immiscible fluids . during the phase separation of the two fluids ,
the particles accumulate at the interface until those are fully jammed . since the simulations performed by stratford et .
utilize a free energy based multiphase lattice boltzmann model , we show in this section that the multicomponent model introduced in this paper is also able to model the formation of a bijel .
we study the temporal development of the system and compare our results with the results of stratford et al .. further , we investigate the influence of the particle concentration , @xmath127 , and @xmath144 on the bijel formation .
the initial conditions for the simulations are as follows : an identical amount of the two fluid species is distributed randomly throughout the system .
the initial positions of the colorless particles ( @xmath145 ) are also chosen at random . in order to keep the system size at manageable @xmath146 lattice units and to be able to simulate a significant number of particles , the particle radius is kept at @xmath112 lattice units in all simulations . the conversion from lattice units to si units of a system containing two identical fluids with the speed of sound ( @xmath147 m/s ) and kinematic visosity ( @xmath148m@xmath149/s ) of water at 20@xmath150c results in a timestep of @xmath151s and a lattice constant of @xmath152 nm .
since we set our particle diameter to 10 lattice units , the physical diameter would be @xmath153 nm and the side length of a cubic simulation volume with @xmath154 lattice units corresponds to @xmath155 nm . the systems presented in this section are simulated for @xmath156 timesteps corresponding to @xmath157 . , @xmath119 and a particle concentration of 20% .
during the first @xmath158 steps of the simulation the average domain size grows from below 5 to its final value of about 31 lattice units . ] as already stated correctly by stratford et al .
, thermal fluctuations have only little effect on the phase separation and bijel formation and can thus be ignored in the simulations @xcite . to analyze the development of structures in the simulated systems ,
we define the averaged time dependent lateral domain size @xmath159 which consists of an average of its components @xmath160 along direction @xmath161 as given by @xmath162 here , @xmath163 is the second order moment of the three - dimensional structure function @xmath164 with respect to the cartesian component @xmath39 , @xmath165 denotes the average in fourier space , weighted by @xmath166 and @xmath89 is the number of nodes of the lattice , @xmath167 the fourier transform of the fluctuations of the order parameter @xmath168 , and @xmath169 is the @xmath39th component of the wave vector @xcite .
the simulations are performed using a @xmath146 lattice , a coupling constant of @xmath99 , an initial fluid density of @xmath119 , a particle volume ratio @xmath170 of 20 percent ( about 6400 particles ) , a particle size of @xmath112 lattice units , and a particle density of 1 , i.e. the particles are slightly heavier than the fluid .
the time development of the average domain size @xmath159 is shown in fig .
[ fig_jamming ] .
the figure clearly shows that the system comes to arrest after a brief period of phase separation . during the first @xmath158 timesteps the average domain size @xmath159 increases from 5 lattice units to about 31 lattice units and stays at that value until the end of the simulation at @xmath171 timesteps
this qualitatively agrees to the results obtained by stratford et al .
however , their simulation does not converge to a fixed domain size , which might be caused by the thermal motion incorporated in their model . while thermal fluctuations are not strong enough to detach the particles from the interface they might cause some local reordering of the particles and therefore support further domain growth .
this effect would be favored by the small particle diameter used in the simulations presented in @xcite since the particle size is of the same order or even smaller than the interface thickness . at @xmath172 through the system studied in fig .
[ fig_jamming ] .
the spots where @xmath173 correspond to the regions occupied by particles .
@xmath174 shows pronounced differences between @xmath175 and @xmath176 , while at timesteps @xmath177 and @xmath178 only minor rearrangements can be observed . ] the arrest of the phase separation process can also be observed by visualizing a 2d cut at @xmath172 of the order parameter as in fig .
[ fig_jamming_colourfield ] or a 3d visualisation of @xmath174 as in fig .
[ fig_jamming_images ] .
the differences between timesteps @xmath175 and @xmath176 are large while the system barely changes between timesteps @xmath179 and @xmath180 .
both visualisations clearly demonstrate the bicontinuouity of the fluid domains . in particular fig .
[ fig_jamming_images ] depicts how the particles get trapped at the fluid - fluid interface and cause the demixing process to stop . [
cols="^,^ " , ] another well known phenomenon that can be observed in mixtures of immiscible fluids and particles are pickering emulsions @xcite . here , the particles are not necessarily equally wettable by the fluids anymore .
also , the ratio of the amount of fluid of different species deviates from 1 .
the result is a system where one phase is continuous while the other forms droplets which are stabilized by the particles .
the particles prevent the droplets from merging when they collide and therefore stop the growth of the average droplet size . as before the droplet size
can be measured utilizing @xmath159 as shown in fig .
[ fig_pickering_c0.15-l ] . here , the lattice is @xmath146 , the interaction constant @xmath181 , and the initial fluid density @xmath182 .
the fluid ratio is 1:3 and the particles with a color of @xmath183 have a volume concentration of 15% . as for the previously presented `` bijels '' a rapid growth of @xmath159 from 5 to over 25 lattice units can be observed during the first 25000 timesteps of the simulation .
the domain growth slows down to a slight decelerating growth afterwards .
this agrees qualitatively with experimental results by arditty et al .
@xcite . through the system described in figure [ fig_pickering_c0.15-l ] at different times .
shown is the order parameter @xmath174 . ]
a three - dimensional visualisation of the order parameter @xmath174 and the particles is shown in fig .
[ fig_pickering_images ] and accompanied by a two dimensional cut of the system at @xmath172 in fig . [ fig_pickering_colourfield ] .
the particle covered droplets as well as the slowing down of droplet growth can be observed . while the system changes dramatically between timesteps 5000 and 10000 almost no difference
can be observed when comparing step @xmath184 to step @xmath185 .
timesteps versus particle concentration @xmath170 .
system size @xmath146 , @xmath99 , @xmath119 , fluid ratio 1:3 and particle color @xmath186 .
increasing @xmath170 leads to a decreasing average domain size . also shown
is a fit with equation @xmath187 .
error bars are given by the maximum deviation of @xmath188 , @xmath189 , @xmath190 from the mean .
] the influence of the interfacial area on the droplet size can be demonstrated by modifying the particle concentration . the resulting average domain size for different concentrations is shown in fig . [ fig_pickering_c ] .
a higher concentration leads to a larger stabilised interfacial area resulting in smaller droplets : reducing the particle concentration from 0.15 to 0.05 corresponds to an increase of the average droplet size from 29 to 41 lattice units .
as the simulated system is finite , modifying the concentration of particles does also change the volume of the two fluid components .
therefore , the inversely proportional relation between particle concentration and droplet size as found by arditty et al .
@xcite does not apply here , but has to be shifted by a constant offset .
@xmath191 is found to be a good fit of the data presented in fig .
[ fig_pickering_c ] .
timesteps versus contact angle @xmath93 .
system size @xmath146 , @xmath99 , @xmath119 , fluid ratio 5:9 and particle concentration @xmath192 .
neutrally wetting particles lead to a small @xmath193 with a large deviation between the spatial directions while strongly colored particles lead to a larger average domain size with almost no deviations .
error bars are given by the maximum deviation of @xmath188 , @xmath189 , @xmath190 from the mean . ] as expected the colour and thus the contact angle of the particle has a drastic influence on the formed structure . while strongly colored particles with contact angles different from 90 degrees lead to spherical droplets , neutrally wetting particles result in droplets that are not as spherical anymore .
we observe structures that are similar to the ones found by kim et al .
@xcite for their simulation of neutrally wetting particles .
these structures are extended in one of the directions .
this results in a reduction of the measured average domain size @xmath193 , while the difference between the directions increases .
this difference is expressed through the error bar in fig .
[ fig_pickering_colour ] .
the values of the contact angle shown in the figure are obtained from the mapping presented in fig .
[ fig_kw_f]b .
in the current section it is demonstrated how the contact angle , the particle volume concentration and the ratio of the two fluid species determine the final state of the system to be a bijel or a pickering emulsion .
phase diagrams depending on the various simulation parameters are presented in fig .
[ fig_pickering_c - c ] and [ fig_pickering_c - ratio ] . in order to reduce the computational cost ,
the size of the lattice has been reduced to 128@xmath194 in this section . however , by performing a small number of 512@xmath194 sized simulations it has been confirmed that finite size effects are still below an acceptable limit and do not influence the final physical state of the system . if the system categorizes as bijel or pickering emulsion
is determined visually after @xmath195 timesteps .
@xmath138 is kept fixed at 0.7 and @xmath196 is set to 0.08 in all simulations .
the system size is @xmath197 , @xmath119 , and the ratio of the two fluid species is kept fixed at 5:9 . for contact angles larger than 90 degrees the system always relaxes towards a bijel ,
while for strongly negative coloring a pickering emulsion is obtained . the line is a guide to the eye.[fig_pickering_c - c ] ] figure [ fig_pickering_c - c ] shows a phase diagram in dependence on the particle concentration and the contact angle ( see fig . [ fig_kw_f]b for the mapping between the particle color and the contact angle ) .
the ratio of the two fluid species is kept fixed at 5:9 . for contact angles larger than 90 degrees the system always relaxes towards a bijel , while for strongly negative coloring and thus smaller contact angles a pickering emulsion is obtained . the particle concentration , however , only has a minor influence on the final state .
it can only be noticed that for small concentrations the formation of a pickering emulsion is more favored for smaller contact angles .
the line is only a guide to the eye since the exact position of the transition from bijel to pickering emulsion would require substantially more data points . in fig .
[ fig_pickering_c - ratio ] the final system state is depicted in dependence on the contact angle and the fluid ratio
. a fluid ratio of at least 3:4 results also for contact angles larger than 90 degrees in a bijel , while for a fluid ratio of 2:5 even neutrally wetting particles are able to stabilize a pickering emulsion . as already shown in fig .
[ fig_pickering_c - c ] for a fluid ratio of 5:9 it depends on the contact angle if the system relaxes towards a bijel or a pickering emulsion .
this behavior can be explained by the interplay between interface curvature and interface size : it is only energetically beneficial if the work required to maintain a curved interface is not larger than the cost due to the increased size of the interface , where the latter can be overcome by a higher concentration of particles at the interface
. timesteps .
the system size is @xmath197 , @xmath119 , and the particle concentration is kept fixed at 0.2 . for fluid ratios between 1:1 and 3:4 also for contact angles larger than 90 degrees
a bijel is obtained . for larger concentration ratios it depends on the particle color or contact angle if a bijel or a pickering emulsion is produced .
the line is a guide to the eye .
[ fig_pickering_c - ratio ] ]
in this paper we proposed a new method allowing the simulation of particles with variable contact angle in multicomponent fluid flows .
we have studied the influence of the model parameters on the resulting fluid - particle interactions and shown that our approach is able to simulate the formation of `` bijels '' and pickering emulsions . by computing phase diagrams we have demonstrated how the transition from bijel to pickering emulsion is determined by the contact angle between particle and fluids , the particle concentration , and the ratio of the two fluid species : while the wettability of the particles and the fluid ratio strongly influence the transition from a bijel to a pickering state , the particle volume concentration only has a minor impact .
we like to thank the dfg for funding within sfb 716 .
further funding is acknowledged from fom ( ipp ipogii ) and nwo / stw ( vidi grant of j. harting ) .
we thank f. janoschek and s. schmieschek for fruitful discussions .
in particular the contribution of f. janoschek to the development of the simulation code is highly acknowledged .
simulations have been performed at the scientific supercomputing centre karlsruhe and the jlich supercomputing center .
a. narvez , t. zauner , f. raischel , r. hilfer , and j. harting . quantitative analysis of numerical estimates for the permeability of porous media from lattice - boltzmann simulations .
, 2010:p211026 , 2010 . | particle stabilized emulsions are ubiquitous in the food and cosmetics industry , but our understanding of the influence of microscopic fluid - particle and particle - particle interactions on the macroscopic rheology is still limited . in this paper
we present a simulation algorithm based on a multicomponent lattice boltzmann model to describe the solvents combined with a molecular dynamics solver for the description of the solved particles .
it is shown that the model allows a wide variation of fluid properties and arbitrary contact angles on the particle surfaces .
we demonstrate its applicability by studying the transition from a `` bicontinuous interfacially jammed emulsion gel '' ( bijel ) to a `` pickering emulsion '' in dependence on the contact angle , the particle concentration , and the ratio of the solvents . |
one of the most celebrated applications of quantum information processing is grover s quantum search algorithm , which allows an initially unknown element to be determined from @xmath0 equally likely possibilities in @xmath1 queries @xcite .
this outperforms the optimum classical strategy ( a random trial and error of elements ) , which requires @xmath2 steps on average .
in addition to providing a speed - up of the unstructured search problem , grover s algorithm can also be adapted to look for solutions to a range of mathematically difficult problems that have structure , by nesting one quantum search inside another @xcite . as with other applications of quantum computing , the benefits of a quantum over a classical approach increase with the size of the database .
indeed , it has been suggested that the _ primary resource _ for quantum computation is a hilbert - space dimension , which grows exponentially with the available physical resources @xcite .
another important consideration is that the physical and temporal resources required to implement the algorithm do not grow too rapidly with the register size @xmath0 .
proof - of - principle quantum search has been successfully demonstrated in nuclear magnetic resonance @xcite , linear optical @xcite and trapped - ion systems @xcite , as well as with individual rydberg atoms @xcite and in classical optics @xcite . of these , only the trapped - ion platform possesses a fully scalable hilbert space and in this sense it is realistically the only candidate for performing a practically useful quantum search .
we note , however , that while the trapped - ion system is scalable @xcite , the largest dimensional quantum search so far performed with trapped ions was for a database size of @xmath3 @xcite . extending the approach of ref .
@xcite to a large number of ions is highly demanding , since it requires the ability to construct , with very high fidelity , a great number of multiply - conditional gates . for two ions , as in ref .
@xcite , each oracle query amounts to a controlled phase gate between the ions . when @xmath4 ions are involved , however , the oracle operator becomes a gate which is multiply - conditional upon the internal state of _ all _ @xmath4 _ ions_. although such a multiply - conditional gate may be decomposed efficiently using a series of one- and two - ion gates @xcite , in practice such a synthesis becomes a daunting task , even for a moderate number of ions .
for example , for a four - qubit register , the oracle may be constructed using 13 two - qubit conditional gates @xcite , each of these requiring 5 consecutive physical interactions on average , which is beyond the capabilities of current experiments @xcite .
the above example makes it apparent that there is a clear distinction between an approach , which is only _ formally scalable _ , and an approach , which is _ realistically scalable _ using current technology .
it is highly desirable , therefore , to find new ways of performing quantum search in a scalable system , which does not require an exorbitant number of elementary physical interactions . in this paper
we propose a novel approach to perform grover s search algorithm .
the linear crystal of @xmath4 ions is prepared in a symmetric dicke state with @xmath5 excitations , @xmath6 ; hence the register dimension scales _ exponentially _ with the number of ions .
the implementation of the quantum search involves a series of red - sideband laser pulses addressing all , or half , of the ions in the trap .
thus each of the inversion about average and the oracle query can be produced in a _
single _ interaction step .
consequently , the total number of _ physical _ steps is the same as the number of algorithmic steps . the implementation of the grover s search algorithm proposed below follows earlier proposals , wherein the database scales linearly @xcite ( see also @xcite ) and quadratically @xcite with the number of ions @xmath4 .
the database size here , which is @xmath7 , scales exponentially with @xmath4 and it is therefore of fully quantum nature .
the reduction of the full hilbert space of dimension @xmath8 by a factor of @xmath9 is compensated by the absence of any ancilla qubits or the need of error correction because of the ultra - compact implementation , in which each logical mathematical step is realized with a single physical interaction ( laser pulse ) .
the remainder of this paper is organized as follows .
a brief review of grover s algorithm for a quantum mechanical speed - up of the unstructured search problem is given in sec .
[ quantumsearch ] .
particular attention is given to the ideas of amplitude amplification and generalized reflections , which lie at the heart of the search algorithm .
[ system ] introduces a model hamiltonian describing a single laser pulse illuminating a chain of trapped ions , and a convenient hilbert - space factorization , with which to describe the dynamics . by making use of the simplified dynamics in this factorized basis ,
we construct recipes for the approximate synthesis of the two reflections required for a quantum search , and examples of numerical simulations are presented in sec .
[ algorithm ] . in sec .
[ conclusions ] we summarize our findings .
grover s algorithm provides a method for solving the unstructured search problem , which can be stated as follows : given a collection of database elements @xmath10 , and an _ oracle function _
@xmath11 that acts differently on one _ marked _
element @xmath12 to all others , @xmath13 find the marked element in as few calls to @xmath11 as possible @xcite .
if the database is encoded in a physical system that behaves classically , then each oracle query can only act on a single database element . in this case , the optimal search strategy is simply a random selection of elements ; on average , it will be necessary to make approximately @xmath14 calls to the oracle before the marked element @xmath12 is located .
the idea underlying grover s algorithm is to encode the database in a physical system that behaves quantum mechanically .
therefore , each possible search outcome is represented as a basis vector @xmath15 in an @xmath0-dimensional hilbert space ; correspondingly , the marked element is encoded by a _ marked state _ @xmath16 .
hence one can apply unitary operations ( involving the oracle function ) to _ superpositions _ of the different search outcomes .
it is thus possible to amplify the amplitude of the marked state @xmath16 using constructive interference , while attenuating all other amplitudes , and locate the marked element in @xmath1 steps . before the execution of the algorithm ,
the quantum register is prepared in an _ equal _ superposition of all basis elements , @xmath17 central to the operation of the quantum search algorithm is the idea of generalized complex reflections , known in the computer science literature as _ householder reflections _
( hr ) @xcite : @xmath18 when the phase @xmath19 is set equal to @xmath20 , the effect of @xmath21 on any vector is to invert the sign of the component of this vector along @xmath22 , while leaving all other components unchanged , which amounts to a reflection with respect to an @xmath23-dimensional plane orthogonal to @xmath22 . allowing @xmath19 to take arbitrary values extends the concept of reflection , by imprinting an arbitrary phase onto the component along @xmath22 , rather than a simple sign inversion .
householder reflections are widely used in classical data analysis and also constitute a powerful tool for coherent manipulation of quantum systems @xcite .
the core component of grover s algorithm is a pair of coupled householder reflections , which together form a single _ grover operator _
@xmath24 : @xmath25 according to standard nomenclature , the operator @xmath26 is referred to as the oracle query , while @xmath27 is known as the inversion - about - average operator .
we note that with the initial state given in eq . , and during successive applications of the operator @xmath24 , the state vector for the system begins and remains in the two - dimensional subspace defined by the non - orthogonal states @xmath16 and @xmath28 .
each application of @xmath24 amplifies the marked state population until it reaches a maximum value close to unity after @xmath29 iterations , at which point the search result can be read out by a measurement in the computational basis . the problem of how to optimize the quantum search routine by allowing arbitrary @xmath30 and
@xmath31 has been studied extensively @xcite .
it is found that the maximum possible amplitude amplification per step of the marked state arises when the phases @xmath30 and @xmath31 are both set to @xmath20 ( as in grover s original proposal @xcite ) .
the corresponding minimum number of search steps @xmath32 is given by : @xmath33 \overset{\mathcal{n}\gg 1}{\sim } \left[\frac{\pi\sqrt{\mathcal{n}}}{4}\right],\ ] ] where @xmath34 $ ] denotes the integer part of @xmath35 .
however , this choice of phases is not unique . for large @xmath0 , as long as the _ phase matching _
condition @xmath36 is satisfied @xcite , a high fidelity search can be performed for any value of @xmath37 in the range @xmath38 and for certain values of @xmath37 , a deterministic quantum search is possible @xcite . in the following two sections , we shall describe how to perform a physically efficient quantum search using the dynamic symmetries in a system of trapped ions .
we consider @xmath4 ions confined and laser cooled in a linear paul trap , each with two relevant internal states @xmath39 and @xmath40 , with respective transition frequency @xmath41 . the linear ion crystal interacts with a laser pulse with frequency @xmath42 , where @xmath43 is the axial trap frequency and @xmath44 is the laser detuning from the first red - sideband resonance .
we assume that the phonon spectrum can be resolved sufficiently well that only the center - of - mass mode is excited by this interaction and that other vibrational modes can safely be neglected . in the lamb - dicke limit , and after making the optical and vibrational rotating - wave approximations , the interaction hamiltonian is @xcite : @xmath45 . \label{hamiltonian}\ ] ] here @xmath46 and @xmath47 are the raising and lowering operators for the internal states of the @xmath48th ion , @xmath49 is the pauli spin matrix and @xmath50 and @xmath51 are respectively the creation and annihilation operators of center - of - mass phonons .
the coupling between the internal and motional degrees of freedom is @xmath52 , where @xmath53 is the single - ion lamb - dicke parameter , with @xmath54 being the laser wave vector and @xmath55 is the mass of the ion .
the function @xmath56 is the real - valued time - dependent rabi frequency .
the hamiltonian conserves the total number of excitations ( @xmath57 ionic plus @xmath58 motional ) , which in the scheme we propose , is half the number of ions , i.e. @xmath5 ( with @xmath4 even ) , @xmath59 .
the energy pattern splits into manifolds corresponding to @xmath60 ionic and @xmath61 motional excitations .
each manifold is @xmath62-fold degenerate , where @xmath63 .
it is readily verified that the dimension of the @xmath64 manifold , @xmath65 , grows _ exponentially _ with @xmath4 ; indeed , for large @xmath4 we have [ database size ] c_n/2^n~ .
the subspace of the overall hilbert space , which spans the manifold @xmath65 , we shall use to represent the state of the register in sec .
[ algorithm ] ; @xmath65 is the set of states , which encode the database of @xmath0 elements . ions with 3 excitations consists of a series of independent chainwise linkages . since the total number of excitations is half the number of ions , then from eq .
follows that states with @xmath66 are inaccessible and are therefore not shown .
the states that make up the longest ladder are all symmetric dicke states @xcite .
the number of motional @xmath67 and ionic @xmath60 excitations for each level can be inferred from @xmath68 , eq . .
] it is possible to move to a new basis in which the hilbert space is factorized into a collection of noninteracting chains of states .
the new states we shall call ms states , since they can be obtained by the multilevel morris - shore ( ms ) factorization @xcite . to determine the ms states first we rewrite the hamiltonian in terms of the total ionic pseudospin : @xmath69 where @xmath70 and @xmath71 .
the ms basis consists of the set of the eigenvectors of the two commuting operators @xmath72 and @xmath73 , where @xmath74 .
therefore each ms state is assigned two quantum numbers , respectively @xmath75 and @xmath68 .
since @xmath72 commutes also with the hamiltonian , the hilbert space factorizes into a set of decoupled chains with different values of @xmath75 .
the ms states are sketched in fig .
[ fig1 ] for 6 ions with 3 excitations . because the laser pulse couples equally to every ion in the trap , the longest chain is comprised of the symmetric dicke states @xmath76 , wherein a given number of ionic excitations @xmath57 is shared evenly amongst all the ions in the trap @xcite .
each chain is assigned different @xmath75 and is comprised of states with different @xmath68 , which varies from 0 to @xmath75 , corresponding to the number of excited ions .
if @xmath60 of the ions are excited , then @xmath77 since the total number of excitations is half the number of ions , then from eq .
it follows that for each chain @xmath68 has a maximum value of 0 , i.e. states with @xmath66 are inaccessible . by analogy with the traditional angular momentum operators
, it follows that the number of states in a chain is equal to @xmath78 ( @xmath75 of the states are not accessible ) , from which @xmath75 can be inferred .
the longest chain corresponds to @xmath79 , the next longest to @xmath80 and so on , and overall , there are @xmath81 $ ] chains of different length in the factorized coupling scheme . for the following analysis , it will be necessary to go further and calculate the couplings in the ms basis .
the coupling between the neighbors @xmath82 and @xmath83 follows immediately from the matrix elements of the operators @xmath50 and @xmath84 , i.e. @xmath85 as illustrated in fig .
[ fig1 ] , there can be many different degenerate ms states with the same values of @xmath86 and @xmath87 .
we label the states with a given @xmath88 and @xmath89 with @xmath90 , where @xmath91 , with @xmath92 .
the manifold @xmath65 , which encodes the database elements , consists of the states , for which the pseudo - angular momentum projection is @xmath93 , eq . .
this requires that the total number of ions @xmath4 is even .
half of these are in state @xmath40 , while the other half are in state @xmath39 ( @xmath94 ) .
the number of elements @xmath0 in the database therefore scales exponentially with the number of ions @xmath4 , eq . . the elements @xmath15 in eq
. , which belong to @xmath65 , can now be written as @xmath95 where the subscript @xmath96 runs over all distinct permutations @xmath97 of the ions internal states .
the initial state @xmath28 is thus a symmetric dicke state @xmath6 of @xmath4 ions sharing @xmath5 excitations , @xmath98 our proposed experimental procedure consists of four simple operations .
( i ) the ions are firstly initialized in the entangled dicke state .
this may be achieved using very simple adiabatic passage techniques , involving either a pair of chirped laser pulses @xcite or two pairs of delayed but overlapping laser pulses @xcite , and using global addressing .
( ii ) synthesis of the inversion - about - average operator is appealingly simple : @xmath99 is a single red - sideband off - resonant laser pulse addressing all ions in the linear chain .
( iii ) the oracle query @xmath100 is also a single red - sideband laser pulse , applied on half of the ions .
after an appropriate number of iterations , the system evolves into the marked state @xmath16 , which can be identified by performing ( iv ) a fluorescence measurement on the entire chain . in most existing proposals for implementing grover s search algorithm using trapped ions
the generation of @xmath99 requires a large number of concatenated physical interaction steps , even for moderate register size @xmath0 .
however , by restricting the dynamics to the subspace of the overall hilbert space in which only half of the ions are excited , this operator becomes possible to synthesize in only a _
single _ interaction step .
this simplification is achieved by taking advantage of the fact that both the hamiltonian and the state @xmath6 are symmetric under exchange of any two ions .
the energies of the ms states do not cross in time so that in the limit @xmath101 the transitions between the ms states vanish due to the effect of adiabatic complete population return @xcite .
each of the ms states acquires a phase shift @xmath102 ( the index @xmath75 corresponds to the eigenvalue of @xmath72 ) and the unitary propagator within the dicke manifold @xmath65 is a product of @xmath103 coupled reflections @xcite @xmath104 with @xmath105 and @xmath106 . for a given value of the coupling , the detuning may be adjusted in order to control the phases @xmath102 .
ideally , these interaction parameters would be chosen such that @xmath107 ( e.g. , @xmath108 ) , while @xmath109 as this would result in @xmath110 being identical to the inversion - about - average operator @xmath111 .
the effect of each oracle query @xmath112 is to imprint a phase shift @xmath113 on the marked state @xmath16 , whilst leaving all other computational basis states unchanged . in general this
can be achieved by a multiply - conditional phase gate upon the internal state of all @xmath4 ions in the trap .
when more than a few ions are involved , this becomes a prohibitively complicated operation , which generally requires an immense number of one- and two - qubit gates @xcite .
however , since we do not work in the whole hilbert space , but rather in the manifold @xmath65 , the oracle operator can be implemented in a much simpler fashion by a _
single _ red - sideband laser pulse , addressing uniformly those @xmath5 ions in the trap , which share the excitation of the marked state .
let s consider an example , when the marked state is @xmath114 .
then we address the first three ions .
since the initial state @xmath6 and the interactions with the laser are symmetric under exchange of the first three ions , and the last three , as well , during the execution of the algorithm the state of the system is a linear combination of the states @xmath115 , @xmath116 ; the marked state is @xmath117 .
because the hamiltonian , which describes the oracle call , does not change the total number of excitations of the first ( and the last ) three ions , i.e. it does not drive the system between @xmath118 and @xmath119 for @xmath120 , in the adiabatic limit the states @xmath121 acquire only phase shifts @xmath122 .
therefore the propagator in the manifold @xmath65 is formally described by the action of @xmath5 coupled reflections , @xmath123 hence , as in the case of the inversion - about - average operator , we need to control @xmath5 phases .
the interaction parameters should be chosen such that @xmath124 , while @xmath125 , as this would result in @xmath126 being identical to the oracle operator @xmath112 .
the phase conditions for the propagators @xmath127 and @xmath128 , derived above , can not be satisfied exactly , since the phases @xmath129 and @xmath122 are overly commensurate .
however one can still perform the algorithm with sufficiently high fidelity @xmath130 here @xmath131 is the state of the system after the completion of the algorithm and @xmath132 is its deviation due to the phases deviation .
fortunately , a fourier expansion reveals that phase deviation of all odd orders leads to the purely imaginary bracket @xmath133 and hence do not contribute to eq . .
as a result , the algorithm is less sensitive to phase deviations since only those of second order have a leading affect on the final state populations .
ions and @xmath5 excitations ( register dimension @xmath134 ) , for @xmath135 ( upper frame ) , @xmath136 ( middle frame ) and @xmath137 ( lower frame ) . the system of ions is assumed to be initialized in the symmetric dicke state @xmath138 .
the laser pulses have a gaussian shape , @xmath139 .
the thin vertical lines display the timing of the oracle call .
the computed numerically scaled detunings and peak rabi frequencies for the oracle and the inversion - about - average operator , respectively , are given in the table below .
the marked - state population ( the fidelity ) of around @xmath140 is obtained after @xmath141 , 6 and 12 steps , respectively , in exact agreement with grover s value .
the numerical simulation includes all off - resonant transitions to states with @xmath142 . ]
[ cols=">,>,>,^,^,^,^ " , ] [ fig2 ] as a demonstration , we have solved the schrdinger equation numerically for a gaussian pulse shape and a constant detuning @xmath44 .
sample results are shown in fig .
[ fig2 ] for @xmath135 , 8 and 10 ions with 3 , 4 and 5 excitations , respectively , which imply databases of @xmath143 , 70 and 252 elements .
the fidelity plotted on the vertical axis is the time - dependent population of the marked state .
the system of ions is assumed to be prepared initially in the dicke superposition @xmath6 of ionic collective states , each of which contains exactly @xmath5 ion qubits in state @xmath40 and @xmath5 in state @xmath39 .
each grover iteration consists of a phase shift for the marked state ( oracle call ) , which amounts a single red - sideband laser pulse , addressing uniformly those @xmath5 ions in the chain , which share the excitation of the marked state , followed by a single red - sideband pulse , which addresses uniformly the entire ion chain ( the inversion - about - average operator ) .
the number of steps , for which the algorithm singles out the marked item with a probability of around @xmath140 , is respectively @xmath141 , 6 and 12 as predicted by eq . .
despite the intense flurry of theoretical and experimental activity in the decade following grover s original proposal for a quantum speed - up of unstructured search , a large discrepancy still remains between the current experimental state - of - the - art and what is required for a practically useful quantum search . to highlight this incongruence
, we note that the only physical system in which a scalable quantum search has been performed is in a chain of trapped ions , and in that experiment , the search - space totalled just four elements . an important intermediate goal on the long road to performing a practically useful quantum search is to demonstrate grover s algorithm in a moderately - sized trapped - ion quantum register .
the standard approach of building up the many - ion grover operator using a network of single- and two - qubit gates is inappropriate for this task , since the required physical resources far exceed those available in today s experiments .
by contrast , in this article , we have proposed to construct the many - ion grover operator using only two single off - resonant laser pulses , with a suitably chosen peak rabi frequencies and detunings , which synthesize the inversion - about - average and oracle operators , each in a single shot .
a simple recipe for synthesizing the inversion - about - average operator was derived by factorizing the overall hilbert space into a series of independent ladders .
the coupling strengths between the ms states were determined solely through a consideration of the angular - momentum structure of the combined ionic pseudospin .
the technique proposed in this paper raises the prospect of demonstrating grover s algorithm in a moderately sized trapped - ion database comprising up to several hundred elements , and which scales exponentially with the number of ions ; this is a necessary step on the path to demonstrating a practically useful quantum search , which remains a long - term goal .
this work is supported by european commission s projects emali and fastquast , the bulgarian science fund grants vu - f-205/06 , d002 - 90/08 , sofia university grant 020/2009 , and elite programme of the landesstiftung - baden - wrttemberg . | we propose a scalable implementation of grover s quantum search algorithm in a trapped - ion quantum information processor .
the system is initialized in an entangled dicke state by using simple adiabatic techniques .
the inversion - about - average and the oracle operators take the form of _ single _ off - resonant laser pulses , addressing , respectively , all and half of the ions in the trap .
this is made possible by utilizing the physical symmetries of the trapped - ion linear crystal .
the physical realization of the algorithm represents a dramatic simplification : each logical iteration ( oracle and inversion about average ) requires only _ two _ physical interaction steps , in contrast to the large number of concatenated gates required by previous approaches .
this does not only facilitate the implementation , but also increases the overall fidelity of the algorithm . |
the dynamics and thermodynamics of chemical reaction networks is a subject that goes back at least to @xcite .
recent years have seen considerable interest in the problem at different levels , from the characterization of their mass - action kinetics @xcite and of their stochastic thermodynamics based on the chemical master equation @xcite , to the analysis of their non - equilibrium steady states ( ness ) @xcite . besides their relevance for fundamental understanding
, these approaches provide an important frame for the study of real biochemical systems , like genome - scale reconstructions of cellular metabolic networks @xcite . the most basic information about these systems is usually encoded in the matrix of stoichiometric coefficients , representing in essence the ( weighted ) topology of the couplings between chemical species and reactions . with uncertainties about kinetic parameters and
reaction or transport mechanisms often preventing large - scale kinetic approaches ( with exceptions like the the metabolism of human erythrocytes @xcite ) , the challenge at the simplest level is that of building stoichiometry - based predictive models of metabolic activity at genome scale .
much information on the organization of reaction fluxes in ness can indeed be obtained from constraint - based models that rely on minimal mass - balance @xcite or stability @xcite assumptions .
such descriptions revolve around pre - defined ( and physically motivated ) sets of local constraints , enforcing for instance mass balance at each metabolite node in the network .
it would be interesting ( and instructive ) to _ derive _ the relevant local constraints as the result of a mathematical analysis of the dynamics taking place on the network , both to further clarify the assumptions behind the models and to highlight their limitations .
this work revisits the joint dynamics of concentrations and reaction fluxes in chemical networks in a statistical mechanics frame . in specific ,
we obtain a variational principle that relates fluxes ( i.e. the average number of microscopic transitions per time per volume for each process ) in ness to the minima of a global function where stoichiometric coefficients and steady - state concentrations appear as parameters .
this function is reminiscent of the hamiltonian of a hopfield model with hebbian couplings @xcite , with stoichiometry playing the role of the patterns .
an analysis of its physical meaning explicitly shows that reaction networks dynamically converge towards states where the use of the available compounds is optimized and the rate of decay of entropy production is minimized .
the flux organization problem turns out to have remarkable similarities with that of optimal resource allocation by heterogeneous agents , as described e.g. by minority games @xcite .
systems of this type generically undergo a transition from an ergodic phase ( the ness is independent of the initial conditions of the dynamics ) to a non - ergodic one when the ratio between the number of reactions and that of chemical species is changed . in our case , the two regimes are described by different sets of local constraints .
we shall first explore this scenario in toy `` random '' reaction networks where such a transition can be fully analyzed numerically .
then a simple real system will be considered , namely the reduced metabolic network model of human erythrocytes .
finally , we shall discuss the relevance of these results for the quantitative analysis of cellular metabolism .
we consider an open system enclosed in a volume @xmath0 ( a reactor or `` cell '' , for brevity ) , formed by @xmath1 distinct chemical species ( labelled @xmath2 ) that can be processed by @xmath3 distinct reactions ( labelled @xmath4 ) at fixed temperature @xmath5 , pressure , ionic strength and ph .
the reaction stoichiometry is described by the coefficients @xmath6 , with the convention that negative ( resp .
positive ) coefficients identify substrates ( resp .
products ) in the ` forward ' direction of reaction @xmath7 .
individual reaction events occur stochastically with rates ( number of events per unit time ) proportional to the substrate concentrations , which vary in time . at stationarity and for ideal systems ,
the gibbs energy ( ge ) change per mole associated to each reaction @xmath7 , i.e. @xcite @xmath8 ( where @xmath9 is the gas constant , @xmath10 the intracellular concentration of species @xmath2 , @xmath11 a reference concentration and @xmath12 the ge change in standard conditions at concentration @xmath11 ) characterizes its distance from detailed balance .
specifically , the forward ( @xmath13 ) and reverse ( @xmath14 ) fluxes of @xmath7 , i.e. the average number of transitions per time per volume , satisfies the relation @xmath15 with @xmath16 @xcite . at equilibrium , @xmath17 and @xmath18 for each @xmath7 .
we focus on the time evolution of the internal composition of such a chemical reactor .
the dynamics of this system is driven essentially by two factors : ( a ) the fact that molecules can stochastically enter or leave the system , and ( b ) the fact that reactions events occur ( stochastically ) inside the system .
reasonably , then , the ness will depend ( a ) on the rate at which molecules can cross the system s boundaries per unit volume ( intake or outtake fluxes , denoted by @xmath19 ) , and ( b ) on the rates of the internal reactions ( more precisely , on the average net number of microscopic transitions per time per volume , denoted by @xmath20 ) . in the following
we will show specifically that , given the boundary fluxes @xmath19 ( characterizing the environment ) , over time scales for which the chemical potentials can be assumed to vary slowly ( so that concentrations can be assumed to be roughly constant ) the internal fluxes in a ness minimize the function @xmath21 ^ 2\geq 0~~,\ ] ] where @xmath22 denotes the concentration of species @xmath2 .
assume that at time @xmath23 the system is characterized by molecular populations ( number of molecules ) @xmath24 .
consider a time interval of size @xmath25 and let @xmath26 denote the net number of transitions of reaction @xmath7 that take place between time @xmath27 and time @xmath28 .
the latter is a random number governed by a probability law ( which shall depend , e.g. , on the concentration of substrates ) that we leave unspecified for the moment .
the corresponding variation of @xmath29 s is given by @xmath30 where @xmath31 stands for the net ( random ) number of molecules of species @xmath2 taken in ( if @xmath32 ) or out ( if @xmath33 ) between time @xmath27 and time @xmath28 .
taking @xmath0 to be fixed , the time - evolution of concentrations @xmath10 is simply @xmath34 we now focus on the quantity @xmath35\ ] ] ( with @xmath36 ) , whose change between times @xmath27 and @xmath28 is given by @xmath37 with initial conditions @xmath38 $ ] .
equation ( [ ix ] ) contains sources of stochasticity in the @xmath39 s and the @xmath40 s . as a consequence , the `` macroscopic '' variables @xmath10 and @xmath41 will fluctuate stochastically as well .
our aim is to characterize the steady state(s ) of ( [ yy ] ) .
we make the following simplifying assumptions : 1 .
molecular populations are large enough to allow us to treat @xmath10 as a continuous variable .
this is generically assumed to be the case in real cells , although e.g. in _
e. coli _ the number of copies of certain small molecules can be as low as a few tens ( corresponding to a concentration of the order of 10 nm @xcite ) .
the effects induced by molecular noise can be non trivial @xcite and accounting for it might alter the emerging picture @xcite .
the quantity @xmath42 is small ( i.e. the chemical potential @xmath43 , with @xmath44 the standard chemical potential , is roughly constant ) , so that @xmath45 can also be taken to be small . under the above assumptions the right - hand side of ( [ yy ] ) is easily linearized to yield @xmath46\nu_j(t)+\frac{1}{v}\sum_{\mu}\frac{\xi_i^\mu b^\mu(t)}{x^\mu(t)}~~.\ ] ] this equation highlights the way in which concentrations affect the time evolution of the system and , in principle
, one would now need to analyze the coupled system formed by ( [ x ] ) and ( [ ydot ] ) .
however , for simplicity , we replace @xmath47 with some time - independent limit value @xmath22 .
this approximation can only be justified as long as one considers evolution over time scales shorter than @xmath48 .
if however concentration changes are sufficiently slow ( in agreement with homeostasis ) it is reasonable to expect that it will hold over time scales much longer than those required to reach a ness . with this , ( [ ydot ] ) takes the form @xmath49 where the `` couplings '' @xmath50 are defined as @xmath51 the steady state of ( [ ydot ] ) can now be obtained straightforwardly by dividing both sides by @xmath25 and averaging over time .
this gives @xmath52 note that @xmath53 is the net flux of reaction @xmath7 ( the average net number of microscopic transitions per time per volume ) while @xmath54 represents the uptake of species @xmath2 ( the average number of molecules of species @xmath2 per time per volume entering or leaving the cell ) . defining @xmath55 we therefore have @xmath56 with @xmath57 ^ 2\geq 0~~.\ ] ] this implies that , on average , the stochastic dynamics of reactions collectively minimizes the function @xmath58 so that the ness of ( [ lang ] ) correspond to the solutions of the minimization problem @xmath59 constrained by the uptake values and by the fact that the @xmath20 s are assumed to be bounded by enzyme availability , so that @xmath60 $ ] .
following @xcite , we can characterize the behavior of fluxes in ness by noting that the solutions of ( [ lang ] ) for @xmath61 ( which are expected to be linked to the gibbs energy change of reaction @xmath7 in a ness ) are generically of the form @xmath62 with @xmath63 , since @xmath64 is constant in ness . in practice
, the stochastic dynamics of @xmath41 for different reactions will be characterized by different values of @xmath64 , depending on the asymptotic behavior . if @xmath41 tends to a finite value as @xmath61 then @xmath65 . recalling that the gibbs energy change is related to the forward - to - reverse flux ratio through the detailed balance condition , this describes the case of a reaction whose microscopic transitions occur bidirectionally even for @xmath61 , i.e. such that the forward and reverse fluxes are both non - zero in the steady state .
specifically , its flux is determined by the condition @xmath66 .
when @xmath67 , instead , @xmath41 increases or decreases linearly in time ( after a transient ) and diverges for @xmath61 so that , in the corresponding ness , reaction @xmath7 will be characterized by unidirectional microscopic transitions for @xmath61 .
note that if @xmath65 then @xmath58 is indeed minimized by solving @xmath68 , whereas for @xmath67 the minimum of @xmath58 is achieved by taking @xmath20 as large ( if @xmath69 ) or as small ( if @xmath70 ) as possible , i.e. @xmath71 ( resp .
@xmath72 ) if @xmath69 ( resp .
@xmath70 ) . for a start ,
notice that the change in @xmath41 is expressed in equation ( [ ydot2 ] ) as the sum of two terms .
the first accounts for the coupling of @xmath7 to other reactions in the network via shared compounds ( the coupling coefficient @xmath50 is non - zero only if @xmath7 and @xmath73 have a metabolite with finite concentration in common ) . consider a species @xmath2 and two reactions @xmath7 and @xmath73 such that @xmath74 ( @xmath73 produces @xmath2 ) and @xmath75 ( @xmath2 is consumed by @xmath7 )
. then @xmath76 and a positive net advancement of reaction @xmath73 will contribute to the increase of @xmath41 , see ( [ ydot2 ] ) .
now looking at ( [ yy ] ) it is reasonable to expect that the probability of observing a forward transition for reaction @xmath7 between time @xmath27 and time @xmath28 will be larger the larger ( and positive ) is @xmath77 ( in agreement with the fluctuation theorem @xcite ; we shall see an explicit example of this later on ) .
therefore in this case a positive net advancement of reaction @xmath73 will ultimately increase the probability of a concomitant advancement of reaction @xmath7 .
in other words , this situation favors the emergence of a positive correlation between reactions @xmath7 and @xmath73 .
likewise , if both @xmath7 and @xmath73 are either producing ( @xmath78 , @xmath74 ) or consuming ( @xmath75 , @xmath79 ) species @xmath2 , their coupling will tend to anti - correlate @xmath39 and @xmath80 .
in this case the dynamics discourages over - production or over - consumption of a chemical species .
a similar picture holds for the coefficients @xmath81 , which are related to the presence of sources ( like nutrients ) and sinks ( e.g. outtakes ) in the network
. a non - zero @xmath81 acts as a force of magnitude proportional to @xmath82 that tends to polarize in a particular direction a reaction @xmath7 that is stoichiometrically connected to a compound @xmath2 with @xmath83 .
the favoured direction depends on whether @xmath2 is a source or a sink and on the sign of @xmath6 .
this effect can propagate to other nodes connected to @xmath7 if @xmath82 is sufficiently large .
the role of the concentrations @xmath22 appearing in @xmath50 and @xmath81 is in essence that of modulating the strength of the couplings and of the forcing fields with the availability of the intermediate metabolites .
indeed , @xmath50 s get stronger when the intermediate compounds are present in smaller amounts , stressing the emergence of positive correlations between processes ( if @xmath76 ) or the limits imposed by competition for a limited resource ( if @xmath84 )
. when the concentration of the intermediate is large , instead , the coupling gets weaker and @xmath7 and @xmath73 may become effectively independent .
the impact of @xmath81 is understood along similar lines . in this way , the original bipartite network of reactions and metabolites can be re - cast as a system of interacting reactions with hebbian couplings , as shown in fig .
[ uno ] . * ( a ) * toy reaction network with reactions represented as circles and chemical species as squares .
continuous , dashed , incoming drawn and outgoing drawn arrows denote stoichiometric coefficients and uptakes , respectively @xmath78 , @xmath75 , @xmath85 and @xmath86 . *
( b ) * reduced reaction network with couplings and `` fields '' given by ( [ jay ] ) . continuous , dotted , incoming grey and outgoing grey arrows denote respectively @xmath76 , @xmath84 , @xmath87 and @xmath88 . for instance , @xmath89 , @xmath90 .
grey arrows are double - headed when the sign of @xmath91 depends on the precise values of stoichiometric coefficients and uptake fluxes .
for instance , the value of @xmath92 depends on the choice of the @xmath93 s and @xmath94 s , since the first term in the sum is negative while the second is positive.,width=302 ] this is strongly reminiscent of hopfield models of neural networks .
in such a scenario , finding the steady state(s ) of ( [ ydot ] ) is equivalent to finding the ground states of a system of reactions interacting with `` energy '' @xmath58 . from a physical standpoint
, @xmath58 quantifies the resource mis - usage by the network so that , by minimizing @xmath58 , the system strives to reach states in which compounds are used as optimally as possible , given the initial conditions @xmath95 ( that also account for the standard ges and hence , to some degree , for the _ a priori _ reversibility ) , the stoichiometry , the available nutrients , the production goals , etc . whether for a _ given network _ the minimum of @xmath58 is zero or
not then depends on several factors , including the bounds on fluxes and the specific form of the uptakes .
note that @xmath96 would imply that at stationarity @xmath97 i.e. that in the steady state a kirchhoff - law type of scenario holds in which strict mass - balance conditions are satisfied for each chemical species .
the network in this case organizes the fluxes so that consumption and production exactly match for each species and meet the nutrient availability and outtake requirements described by the vector @xmath98 . on the other hand ,
the physically relevant states with @xmath99 can more generally be thought to have @xmath100 otherwise the system could e.g. consume a nutrient in excess of its availability . both sets of conditions have been employed for the modeling of cellular metabolic networks . in particular equations of the type of ( [ mb ] )
are the basis of the highly successful flux - balance - analysis ( fba ) @xcite , where biological functionality is included as an additional _ ad hoc _ constraint usually represented by the maximization of a specific score function ( e.g. biomass production for _ e. coli _ in an optimal environment ) .
the conditions ( [ vn ] ) are instead reminiscent of von neumann s model of reaction networks @xcite , where self - consistent flux states with a net positive production of intracellular metabolites can be allowed
. this type of approach can be helpful in analyzing the metabolic capabilities of an organism and , if statistically robust production profiles emerge , in inferring ( rather than postulating ) cellular objective functions , with the idea that chemical species that are globally produced ( e.g. amino acids ) are employed in macromolecular processes ( like protein formation ) that are not encoded by the reaction stoichiometry .
we remark however that within the above setting finding the ness of the system means minimizing @xmath58 , which obviously is a priori different from solving ( [ mb ] ) or ( [ vn ] ) .
a broader physical insight can be obtained by studying how @xmath58 relates to standard quantities used to characterize non - equilibrium behavior in reaction networks , such as the entropy production .
indeed the entropy production per volume @xmath101 of the system enclosed in the `` cell '' of volume @xmath0 is given by @xmath102 where @xmath10 is the concentration of species @xmath2 and @xmath43 is its chemical potential .
taking time derivatives keeping in mind that at stationarity @xmath103 is constant , one easily finds that @xmath104 .
now since @xmath105 , @xmath106 .
in addition , @xmath58 is constant in ness .
it follows that for time scales shorter than @xmath107 ( i.e. for time scales over which chemical potentials are roughly constant ) one has @xmath108 @xmath58 is thus seen to play the role of the rate at which the entropy production changes in a ness . if @xmath96 ( i.e. if the ness is flux - balanced ) , then @xmath101 must vanish as well ( since @xmath109 ) , leading to a state with constant entropy .
if @xmath110 , instead , the entropy production is non - zero and decreases at the smallest allowed rate . correspondingly , the entropy `` slows down '' quadratically over the time scales for which the theory holds .
( note that for sufficiently long times the entropy production can become negative if @xmath111 , implying that this limit may be unphysical . ) in summary , the ness that can be obtained are either characterized by @xmath112 , zero entropy production and hence constant entropy , or by @xmath111 , positive entropy production decreasing in time as slowly as possible and entropy increasing in time accordingly .
it is noteworthy that this scenario essentially characterizes the lyapunov condition described in @xcite for the stability of stationary states ( chapter 18 ) .
the lyapunov function in our case can be computed explicitly , takes the form ( [ h ] ) , its minimization bears a further physical meaning in terms of optimal resource allocation , and provides an equivalent description of the reactor as a system of processes interacting via hebbian couplings .
indeed the quantity @xmath58 appears in the gibbs theory of thermodynamic stability for ideal systems .
let us consider a system at equilibrium with vanishing net fluxes and evaluate the effect of a perturbation that drives the system away from equilibrium .
such a perturbation corresponds to a ( small ) change in the turnover @xmath113 of each reaction @xmath7 , which can be achieved e.g. by forcing a ( small ) non - zero flux @xmath20 through each reaction @xmath7 for a time @xmath25 , so that @xmath114 .
the free energy per volume associated to the perturbed state is easily found to be given by @xmath115 in agreement with the second law of thermodynamics . after turning off the perturbation
, the system will relax back to equilibrium by minimizing @xmath116 ( i.e. @xmath58 ) .
a simple algebraic argument allows to understand that , generically , a qualitative change in the solutions of ( [ min ] ) is expected to take place as one varies the ratio between the network parameters @xmath3 and @xmath1 .
let @xmath117 denote the number of reactions that remain asymptotically bidirectional , i.e. such that @xmath65 .
the conditions @xmath118 that they must satisfy form a set of @xmath117 equations , @xmath1 of which at most are independent ( strictly speaking , the rank of the matrix @xmath119 equals that of the stoichiometric matrix @xmath120 , see ( [ jay ] ) ; it is however always possible to eliminate dependencies in the latter so as to attain full rows rank ) .
neglecting the fact that variables are bounded , we can say that when the number of variables ( @xmath117 ) exceeds that of equations the system is underconstrained and multiple solutions occur .
so multiple steady states exist when @xmath121 .
note that @xmath117 is determined _ dynamically _ by ( [ ydot ] ) with initial conditions @xmath95 .
this suggests that different choices of @xmath95 will lead to different steady states , i.e. that when @xmath121 ergodicity ( i.e. independence of the steady state on initial conditions ) wo nt hold .
a transition is thus expected to occur when @xmath122 . such a transition can be fully investigated within the following , simple model : the stoichiometric coefficients @xmath120 are independent , identically distributed random numbers ( such that each reaction uses and produces a finite fraction of the possible compounds ) , and the dynamics of the network advances in discrete time steps of size @xmath25 . in specific , at each time @xmath27 the @xmath39 s take on values stochastically in @xmath123 with @xmath124 i.e. with @xmath125 $ ] .
note that the above probability ratio is proportional to the ratio between the concentration of substrates and that of products of the reaction .
clearly , ( [ db ] ) does not allow to capture the temporal structure induced by the arrhenius law , because by forcing each reaction to take place at every time step it neglects the fact that activation energies ( and hence characteristic timescales ) can differ significantly across reactions .
it is however reasonable to think that the steady state will be unaffected by these transients . on the other hand ( [ db ] )
makes the theory considerably easier from a mathematical viewpoint and the final result for the steady state ( which is the focus of the present work ) more transparent . setting @xmath126 for simplicity , so that @xmath127 for each @xmath7 , one can follow @xcite to derive the continuous - time limit of ( [ ydot ] ) for large @xmath3 and @xmath1 .
the result is the langevin process @xmath128 where @xmath129 is a gaussian noise with zero mean .
clearly , time - averaging leads back to ( [ lang ] ) with @xmath130 .
recalling that @xmath15 and noting that this implies @xmath131 where , by ( [ db ] ) , @xmath132 , this in turn suggests the relation @xmath133 which explicitly links the thermodynamic driving force of a reaction to the ( stochastic ) dynamics of the quantity @xmath41 .
the fact that the final result differs from the nave intuition @xmath134 is a direct consequence of the stochastic fluctuations encoded in ( [ db ] ) .
note that the dynamics ( [ db ] ) thus converges to ness ( minima of @xmath58 ) that are thermodynamically feasible , in agreement with the second law of thermodynamics .
we can study the linear stability of ( [ langu ] ) by setting @xmath135 where @xmath136 is a ness and @xmath137 is a zero - average noise representing ( small ) perturbations to the trajectory .
reactions with @xmath138 , for which @xmath41 diverges , will be insensitive to @xmath137 .
hence it suffices to focus our attention on the response of reactions for which @xmath139 . to first order in @xmath140
, fluctuations are easily found to obey the condition @xmath141~~,\end{gathered}\ ] ] where @xmath142 .
now if all eigenvalues of the matrix @xmath143 are positive the dynamical system will be linearly stable as small perturbations occurred along the trajectories will die out in time .
the term @xmath144 is clearly positive , hence it suffices to check that the smallest eigenvalue of the matrix @xmath145 is positive .
assuming that @xmath6 are independently and identically distributed , the spectrum of @xmath50 can be computed , in the limit @xmath146 with @xmath147 finite , using the results of @xcite . for the smallest eigenvalue one finds @xmath148 $ ] , where @xmath149 and @xmath150 is a constant .
hence stability ( and ergodicity ) requires @xmath151 .
when @xmath152 , instead , the dynamics will be sensitive to small perturbations and , reasonably , its steady state will be selected by the initial conditions @xmath95 .
the marginal stability condition @xmath153 coincides with the rough algebraic estimate of the transition point made above , i.e. @xmath122 . to illustrate the scenario underpinned by the above theory , we have simulated the dynamics defined by @xmath154 for an ensemble of artificial reactors formed by @xmath1 species and @xmath3 reactions in order to analyze the dependence of its steady state(s ) on @xmath147 .
stoichiometric coefficients were chosen randomly , so that @xmath155 with probability @xmath156 and @xmath157 with probabilities @xmath158 , independently on @xmath7 and @xmath2 .
similarly , for the boundary metabolites we took @xmath159 with probability @xmath160 and @xmath161 with probability @xmath162 , independently on @xmath2 . for sakes of simplicity , we set @xmath163 for all @xmath2 ( it is clear that the choice of @xmath22 does not affect the transition point where @xmath58 vanishes ; it only changes the value of @xmath164 in the ergodic phase ) .
no prior assumption on reversibility was made , i.e. all microscopic transitions can initially occur in both directions for each process .
the coefficients @xmath50 and @xmath81 are defined as in ( [ jay ] ) and ( [ acca ] ) except for the fact that @xmath50 is re - scaled by @xmath165 to ensure that the system is well - behaved when @xmath166 , while @xmath26 evolves according to ( [ db ] ) . for consistency ,
@xmath58 for this case is defined as @xmath167 ^ 2~~.\ ] ] results for @xmath168 are shown in figure [ due ] .
* figure 2 . * average stationary values of @xmath58 , fraction of asymptotically unidirectional reactions ( @xmath169 ) and relative number of asymptotically bidirectional reactions @xmath149 versus @xmath147 obtained from ( [ dynam ] ) ( with no prior assumption on reaction reversibility for an ensemble of random reaction networks with @xmath170 constructed as described in the text .
averages are taken over 200 realizations for each value of @xmath171 . unbiased i.c .
( initial conditions ) refers to steady states of ( [ dynam ] ) for @xmath172 ; biased i.c .
instead correspond to @xmath173 for each @xmath7 .
note the two phases with @xmath111 ( @xmath174 and @xmath112 ( @xmath175 ) as predicted .
the critical point @xmath176 coincides within numerical error with the point where @xmath177 .
finally , the phase with @xmath112 is non - ergodic : different initial conditions lead to different ness , characterized by different values of @xmath169.,width=302 ] one sees that initializing the dynamics from `` equilibrium '' conditions @xmath172 for each @xmath7 one ends up in stationary states with @xmath111 for @xmath178 whereas @xmath112 for @xmath175 .
similarly , @xmath151 for @xmath179 while @xmath152 for @xmath175 , so that the critical point is indeed marked by the condition @xmath153 .
when the dynamics starts from @xmath173 for each @xmath7 , instead , the system ends up in a different steady state for @xmath175 , as signaled by the different value of the fraction of asymptotically unidirectional processes at stationarity , @xmath180 . in the phase characterized by @xmath111
, the steady state is unchanged .
these results fully confirm the theoretical predictions derived above .
we also note that ( data not shown ) , if @xmath159 for each @xmath2 , i.e. if there is no exchange with the surroundings , the corresponding networks converge to a steady state in which @xmath181 for each @xmath7 , @xmath112 and @xmath182 for each @xmath171 , i.e. to chemical equilibrium . in other words ,
expectedly , boundary fluxes induce ness .
we now turn to a somewhat more realistic case in which nevertheless a full analytic study of the scenario underlying the minimization of @xmath58 is possible , namely the standard reduced model of the hrbc metabolism , which includes the glycolytic and the pentose phosphate pathway only ( 21 reactions among 30 metabolites plus two ionic pumps , namely atpase and nadphase ; see tables i and ii for the numerical details and the network structure used here ) .
they operate by consuming glucose ( glc ) in order to , respectively , maintain the osmotic balance through the sodium - potassium ionic pump ( atpase ) , and reduce the amount of free radicals through glutathione reductase ( nadphase ) . moreover , at a simplified level , one can think that the final products of these pathways are , respectively , lactate ( lac ) plus an exchange of k@xmath183 and na@xmath183 ions , and co@xmath184 .
the partitioning of glc between the two pathways depends on the level of oxidative stress faced by the cell .
experimental estimates based on enzyme activities range from 70% or more in favor of glycolysis in unstressed conditions to 70% or more in favor of the pentose - phosphate pathway under oxidative stress @xcite .
we want to use the theory described here to estimate the range of variability of the fraction of glucose consumed by each pathway as a function of the oxygen concentration in the environment .
we start by assuming ( reasonably ) that the steady state of the hrbc metabolism is compatible with @xmath96 ( i.e. flux balance ) .
a straightforward analysis of the emerging equations reveal that only three of the 23 reactions are linearly independent : we choose the glucose uptake @xmath185 , the flux through the rapoport - leubering shunt ( or through the enzyme 2,3-dpg mutase , dpgm , a key step that regulates the haemoglobin s affinity with oxygen ) @xmath186 and the flux through the pentose phosphate pathway ( or through the enzyme glucose-6-phosphate dehydrogenase , g6pdh ) @xmath187 .
all fluxes can be written in terms of these .
in particular , one finds @xmath188 now the variation of the _ extracellular _ concentrations of glc , lac , k@xmath183 , na@xmath183 and co@xmath184 due to the operation of a single hrbc is easily seen to be given by @xmath189 in turn , for the extracellular medium one has @xmath190 minimizing this at fixed @xmath185 and @xmath191 one finds that @xmath192 where @xmath193 and @xmath194 are defined respectively as @xmath195 one sees that if the concentration of co@xmath184 is much larger than the others ( implying @xmath196 ) and @xmath197 then @xmath198 so that the pentose phosphate pathway consumes roughly all of the glucose ( the factor 6 is in agreement with the stoichiometry of carbon atoms in glc and co@xmath184 ) .
we can therefore define the fraction of glc consumption through the ppp by re - scaling @xmath199 by @xmath200 : @xmath201 from this it is also immediately clear that , in general conditions for co@xmath184 , @xmath202 varies between @xmath203 ( corresponding to @xmath197 ) and @xmath204 ( corresponding to maximal @xmath205 ) . using the typical concentration values for the above metabolites in the blood , namely @xmath206 m , @xmath207 m
, @xmath208 m and @xmath209 m one has @xmath210 and @xmath211 so that @xmath212 . despite the roughness of the network reconstruction we employed , the bounds just obtained are in remarkable agreement with experimental evidence . on the other hand , using the empirical estimates
@xmath213 m/s and @xmath214 m/s we find @xmath215 .
in this paper we have derived a variational principle for the ness of chemical reaction networks , showing that , for time scales over which chemical potentials can be considered constant , stationary non - equilibrium fluxes minimize the function @xmath58 , see ( [ h ] ) , in which stoichiometry , intakes , outtakes and concentrations appear as parameters . the cost function is closely related to hopfield models of neural networks .
this allows to rephrase such networks as systems of reactions interacting via hebb - like rules and subject to an external forcing provided by the boundary fluxes .
furthermore @xmath58 bears two simple physical interpretations . first , minimizing it amounts to finding the flux organizations that keep the overall `` waste '' of chemical species to a minimum , given the boundary conditions to be satisfied .
second , it equals ( modulo a constant ) the time derivative of the entropy production per volume : ness thus correspond to flux configurations such that the entropy production decreases in time at the smallest allowed rate , in line with the ideas exposed in @xcite .
we have investigated the implications of such a picture in toy ( random ) chemical networks where the dependence of the solutions on the network s structural parameters can be fully explored , revealing the existence of a phase transition between flux balanced states with @xmath96 and states where unbalances emerge and in a small real biochemical network , namely the metabolic network of the human red blood cell , a system that is possibly the closest to the theoretical situation described here . making contact with stoichiometric models of metabolic networks is relatively straightforward as long as boundary fluxes are taken to be fixed and one does not include additional objective functions that ness are required to maximize .
for instance , the maximization of biomass flux ( a frequent optimization criterion for bacterial metabolism @xcite ) provides a source of entropy production even if one focuses on states with @xmath112 ( as in flux - balance - analysis , fba ) . on the other hand
the theory developed here suggests that the set of local constraints that describes ness is provided by the minimization of @xmath58 rather than by taking @xmath112 or @xmath105 a priori .
an important property that ness should possess is thermodynamic feasibility , i.e. they should not contain infeasible loops @xcite . inspired by the fluctuation theorem @xcite , we have proposed here a simple dynamical rule ( see ( [ db ] ) ) that ensures that the ness obtained as the minima of h are indeed void of cycles . in general
( i.e. when straightforward minimization of @xmath58 is carried out ) , it is possible to get rid of infeasible cycles by complementing the variational problem described here with the minimization of the square norm of the flux vector .
this is a consequence of the gordan theorem of alternatives @xcite : assuming that the matrix @xmath216 has full rank , only one of the following systems has a non - trivial solution : ( a ) @xmath217 ( for @xmath218 real ) ; ( b ) @xmath219 ( for @xmath220 ) . for a reaction network with stoichiometric coefficients @xmath6 and
fluxes @xmath20 , defining @xmath221 one sees that system ( a ) expresses the condition of thermodynamic stability @xmath222 , whereas system ( b ) defines thermodynamically infeasible cycles , so that either a flux configuration @xmath223 is thermodynamically feasible or it contains at least one cycle . now let us consider a steady - state flux configuration satisfying @xmath224 and let us assume it is thermodynamically infeasible , i.e. that system ( b ) has a solution @xmath225 . we can then construct a new flux configuration @xmath226 as @xmath227 , with @xmath228 a constant . evidently , still @xmath229 . however defining @xmath230 ,
it is easily seen that choosing @xmath228 so that @xmath231 one gets @xmath232 . in other terms , starting from a thermodynamically infeasible steady - state flux configuration one can construct another steady - state flux configuration whose total flux is lower . in turn , @xmath233 has to be minimum when the flux configuration is thermodynamically feasible
. it would be interesting to find a dynamical justification of this criterion .
( notice that flux minimization is an optimality principle frequently used for metabolic network modeling , see e.g. @xcite . ) perhaps not too surprisingly , the formal aspects of this theory present many common traits with those developed over the past decade for the analysis of large games with heterogeneous interacting agents , specifically with minority games @xcite . for obvious reasons ,
such an analogy should nt be stretched and we do not elaborate in detail on it here .
in short , however , reactions ( or , more properly , enzymes ) in chemical networks can be thought to be involved in a competition for the use of a set of possibly limited resources ( the different substrates ) . whether a reaction can operate or not depends on how much substrate is available to it , i.e. on the overall substrate concentration and on how many other enzymes can bind the same substrates .
each reaction disposes of two ` strategies ' ( or ways to access the set of resources ) , corresponding to the vectors of its input and output metabolites in the forward and reverse direction , respectively .
such strategies are anti - correlated : substrates in the forward direction are products in the reverse , and vice - versa
. rules like ( [ dynam ] ) can then be read as ` learning ' processes through which enzymes try to anticipate at each time step whether the substrates needed for the forward or reverse processes are most likely to be available , in order for it to operate .
this parallel provides an elementary quantitative flavor to the idea that enzymes in biochemical reaction networks compete for the substrates . in absence of boundary fluxes ( @xmath234 for each @xmath7 in ( [ ydot ] ) ) , the situation is completely equivalent to the minority game with anti - correlated strategies studied in @xcite . in this case
, the dynamics converges to @xmath112 and @xmath181 for each @xmath7 , i.e. the system asymptotically reaches chemical equilibrium . as it should be , ness are induced by non - zero boundary fluxes , i.e. by non - zero fields @xmath81 .
this additional term turns out to be the main difference between standard anti - correlated minority games and the systems discussed here .
besides a possible theoretical interest in deepening the analogy just described ( e.g. by extending the dynamical approaches employed for the analysis of multi - agent systems @xcite to models of chemical reaction networks ) , it will be interesting to see how well the variational principle ( [ min ] ) describes flux states in real biochemical networks .
it is a pleasure to thank e. aurell , w. bialek , e. marinari , m. marsili , i. perez castillo and d. segr for stimulating discussions and suggestions .
adm wishes to thank the kavli institute of theoretical physics china [ project of knowledge innovation program ( pkip ) of chinese academy of sciences , grant no .
kjcx2.yw.w10 ] , the initiative for theoretical sciences at the graduate center , city university of new york , and the abdus salam international centre for theoretical physics for hospitality during the final stages of this work .
this work is supported by the seed project dream of the italian institute of technology ( iit ) and by the joint iit / sapienza nanomedicine lab .
the iit platform computation is gratefully acknowledged .
gu is supported by the firb grant rbfr08ekev .
m. imielinski , c. belta , h. rubin , and a. halasz ( 2006 ) systematic analysis of conservation relations in escherichia coli genome - scale metabolic network reveals novel growth media .
j. * 90 * 2659 c. martelli , a. de martino , e. marinari , m. marsili , and i. p. castillo ( 2009 ) identifying essential genes in escherichia coli from a metabolic .
optimization principle .
usa * 106 * 2607 a. de martino and d. granata and e. marinari and c. martelli and v. van kerrebroeck ( 2010 ) optimal fluxes , reaction replaceability , and response to enzymopathies in the human red blood cell .
j. biomed .
* 2010 * 415148 r.m .
fleming , c.m .
maes , m.a .
saunders , y. ye , b. .
palsson ( 2012 ) a variational principle for computing nonequilibrium fluxes and potentials in genome - scale biochemical networks .
. biol . * 292 * 71 | within a fully microscopic setting , we derive a variational principle for the non - equilibrium steady states of chemical reaction networks , valid for time - scales over which chemical potentials can be taken to be slowly varying : at stationarity the system minimizes a global function of the reaction fluxes with the form of a hopfield hamiltonian with hebbian couplings , that is explicitly seen to correspond to the rate of decay of entropy production over time . guided by this analogy ,
we show that reaction networks can be formally re - cast as systems of interacting reactions that optimize the use of the available compounds by competing for substrates , akin to agents competing for a limited resource in an optimal allocation problem . as an illustration , we analyze the scenario that emerges in two simple cases : that of toy ( random ) reaction networks and that of a metabolic network model of the human red blood cell .
= 1 |
mab problems @xcite are a class of resource allocation problems in which a decision - maker allocates a single resource by sequentially choosing one among a set of competing alternative options called arms . in the so - called stationary mab problem , a decision - maker at each discrete time instant chooses an arm and
collects a reward drawn from an unknown stationary probability distribution associated with the selected arm .
the objective of the decision - maker is to maximize the total expected reward aggregated over the sequential allocation process .
these problems capture the fundamental trade - off between exploration ( collecting more information to reduce uncertainty ) and exploitation ( using the current information to maximize the immediate reward ) , and they model a variety of robotic missions including search and surveillance .
recently , there has been significant interest in bayesian algorithms for the mab problem @xcite .
bayesian methods are attractive because they allow for incorporating prior knowledge and spatial structure of the problem through the prior in the inference process . in this paper
, we investigate the influence of the prior on the performance of a bayesian algorithm for the mab problem with gaussian rewards .
mab problems became popular following the seminal paper by robbins @xcite and gathered interest in diverse areas including controls @xcite , robotics @xcite , machine learning @xcite , economics @xcite , ecology @xcite , and neuroscience @xcite .
much recent work on mab problems focuses on a quantity termed _ cumulative expected regret_. the cumulative expected regret of a sequence of decisions is the cumulative difference between the expected reward of the options chosen and the maximum possible expected reward . in a ground - breaking work , lai and
robbins @xcite established a logarithmic lower bound on the expected number of times a sub - optimal arm needs to be sampled by an optimal policy in a frequentist setting , thereby showing that cumulative expected regret is bounded below by a logarithmic function of time .
their work established the best possible performance of any solution to the standard mab problem .
they also developed an algorithm based on an upper confidence bound on estimated reward and showed that this algorithm achieves the performance bound asymptotically . in the following , we use the phrase _ logarithmic regret _ to refer to cumulative
expected regret being bounded above by a logarithmic function of time , i.e. , having the same order of growth rate as the optimal solution . in the context of the bounded mab problem , i.e. , the mab problem in which the reward is sampled from a distribution with a bounded support , auer @xcite developed upper confidence bound - based algorithms that achieve logarithmic regret uniformly in time ; see @xcite for an extensive survey of upper confidence bound - based algorithms .
bayesian approaches to the mab problem have also been considered .
srinivas @xcite developed asymptotically optimal upper confidence bound - based algorithms for gaussian process optimization .
agrawal and goyal @xcite showed that a bayesian algorithm known as thompson sampling @xcite is near - optimal for binary bandits with a uniform prior .
liu and li @xcite characterize the sensitivity of the performance of thompson sampling to the assumptions on prior .
kaufman @xcite developed a generic bayesian upper confidence bound - based algorithm and established its optimality for binary bandits with a uniform prior .
reverdy @xcite studied the bayesian algorithm proposed in @xcite in the case of correlated gaussian rewards and analyzed its performance for uninformative priors .
they called this algorithm the upper credible limit ( ucl ) algorithm and showed that the ucl algorithm models human decision - making in the spatially - embedded mab problem .
we define a spatially - embedded mab problem as an mab problem in which the arms are embedded in a metric space and the correlation coefficient between arms is a function of distance between them . for example , in the problem of spatial search over an uncertain distributed resource field , patches in the environment can be modeled as spatially located alternatives and the spatial structure of the resource distribution as a prior on the spatially correlated reward .
this is an example of a _ spatially - embedded mab problem . _
it was observed in @xcite that good assumptions on the correlation structure result in significant improvement of the performance of the ucl algorithm , and these assumptions can successfully account for the better performance of human subjects . in this note
we rigorously study the influence of the assumptions in the prior on the performance of the ucl algorithm for a mab problem with gaussian rewards .
since the ucl algorithm models human decision - making well , the results in this paper help us identify the set of parameters in the prior that explain the individual differences in performance of human subjects .
the major contributions of this work are twofold : first , we study the ucl algorithm with uncorrelated informative prior and characterize its performance .
we illuminate the opposing influences of the degree of confidence of a prior and the magnitude of its inaccuracy , i.e. , the gap between its mean prediction and the true mean reward value , on the decision - making performance .
second , we propose and study a new correlated ucl algorithm with correlated informative prior and characterize its performance .
we show that large correlation scales reduce the number of steps required to explore the surface .
we then show that incorrectly assumed large correlation scales may lead to a much higher number of selections of suboptimal arms than suggested by the lai - robbins bound .
this analysis provides insight into the structure of good priors in the context of explore - exploit problems .
the remainder of the paper is organized in the following way . in section [ sec : review ] , we recall the mab problem and an associated bayesian algorithm , ucl . we analyze the ucl algorithm for uncorrelated informative prior and correlated informative prior in section [ sec : uncorr - ucl ] and [ sec : corr - ucl ] , respectively .
we illustrate our results with some numerical examples in section [ sec : numerics ] , and we conclude in section [ sec : conclusions ] .
in this section we recall the mab problem and the bayes - ucb algorithm proposed in @xcite .
the @xmath0-armed bandit problem refers to the choice among @xmath0 options that a decision - making agent should make to maximize the cumulative expected reward .
the agent collects reward @xmath1 by choosing arm @xmath2 at each time @xmath3 , where @xmath4 is the horizon length for the sequential decision process . in the so - called stationary mab problem ,
the reward from option @xmath5 is sampled from a stationary distribution @xmath6 and has an unknown mean @xmath7 .
the decision - maker s objective is to maximize the cumulative expected reward @xmath8 by selecting a sequence of arms @xmath9 .
equivalently , defining @xmath10 and @xmath11 as the expected _ regret _ at time @xmath12 , the objective can be formulated as minimizing the cumulative expected regret defined by @xmath13}}= \sum_{i=1}^n \delta_i { \ensuremath{\mathbb{e}\left [ n_{i}(t ) \right]}},\end{aligned}\ ] ] where @xmath14 is the total number of times option @xmath15 has been chosen until time @xmath16 and @xmath17 is the expected regret due to picking arm @xmath15 instead of arm @xmath18 .
the bayes - ucb algorithm for the stationary @xmath0-armed bandit problem was proposed in @xcite .
the bayes - ucb algorithm at each time 1 . computes the posterior distribution of the mean reward at each arm ; 2 . computes a @xmath19 upper credible limit for each arm ; 3 . selects the arm with highest upper credible limit . in step ( ii )
, the upper credible limit is defined as the least upper bound to the upper credible set , and the function @xmath20 is tuned to achieve efficient performance . in the context of bernoulli rewards , kaufmann @xcite set @xmath21 , for some @xmath22 , and show that for @xmath23 and uninformative priors , the bayes - ucb algorithm achieves the optimal performance .
reverdy @xcite studied the bayes - ucb algorithm in the context of gaussian rewards with known variances . for simplicity the algorithm in @xcite
is called the ucl ( upper credible limit ) algorithm .
it is shown that for an uninformative prior , the ucl algorithm is order - optimal , i.e. , it achieves cumulative expected regret that is within a constant factor of that suggested by the lai - robbins bound .
it is also shown that a variation of the ucl algorithm models human decision - making in an mab task .
in this paper , we focus on the gaussian mab problem , i.e. , the reward distribution @xmath6 is gaussian with mean @xmath24 and variance @xmath25 . the variance @xmath25 is assumed known , e.g. , from previous observations or known characteristics of the reward generation process .
we now recall the ucl algorithm and analyze its performance for a general prior .
suppose the prior on the mean rewards at each arm is a gaussian random variable with mean vector @xmath26 and variance @xmath27 . for the above mab problem ,
let the number of times arm @xmath15 has been selected until time @xmath12 be denoted by @xmath28 .
let the empirical mean of the rewards from arm @xmath15 until time @xmath12 be @xmath29 .
then , the posterior distribution at time @xmath12 of the mean reward at arm @xmath15 has mean and variance @xmath30 respectively , where @xmath31 .
moreover , @xmath32 & = \frac{\delta^2 \mu_{i}^0 + n_i(t ) m_i}{\delta^2+n_{i}(t ) } \ ; \text{and}\ ; \text{var}[\mu_{i}(t ) ] = \frac { n_i(t ) \sigma_s^2}{(\delta^2+n_{i}(t))^2}.\end{aligned}\ ] ] the ucl algorithm for the gaussian mab problem , at each decision instance @xmath3 , selects an arm with the maximum @xmath33-upper credible limit , i.e. , it selects an arm @xmath34 , where @xmath35 @xmath36 is the inverse cumulative distribution function for the standard gaussian random variable , @xmath37 , and @xmath38 and @xmath39 are tunable parameters . in the context of gaussian rewards
, the function @xmath40 decomposes into two terms corresponding to the estimate of the mean reward and the associated variance .
this makes the ucl algorithm amenable to an analysis akin to the analysis for ucb1 @xcite .
using such an analysis , it was shown in @xcite that the ucl algorithm with an uninformative prior and parameter values @xmath41 and @xmath42 achieves an order - optimal performance . in the following , we investigate the performance of the ucl algorithm for general priors . to analyze the regret of the ucl algorithm , we require some inequalities that we recall in the following lemma . [
lem : ineq ] for the standard normal random variable @xmath43 and the associated inverse cumulative distribution function @xmath44 , the following statements hold : 1 . for any @xmath45 @xmath46 2 . for any @xmath47 $ ] , @xmath48 and @xmath49 , @xmath50 statement ( i ) in lemma [
lem : ineq ] can be found in @xcite . the first inequality in ( ii ) follows from ( i ) .
the second inequality in ( ii ) was established in @xcite , and the last inequality can be easily verified using the second inequality in ( ii ) .
[ lem : diff - of - squares ] for any @xmath51 such that @xmath52 , @xmath53 the inequality follows trivially using a completing the square argument .
let @xmath54 , for each @xmath55 .
set @xmath56 , @xmath57 , and @xmath58 , for some @xmath59 .
[ thm : uncorr - regret ] for the gaussian mab problem , and the ucl algorithm with uncorrelated prior , the expected number of times a suboptimal arm @xmath15 is selected satisfies @xmath60 \le \eta_i + \hat n_i(t),\ ] ] where @xmath61 , and @xmath62 is defined in .
@xmath63 ' '' '' see [ proof - uncorr - regret ] .
[ rem : uncor - regret ]
in this section , we study a new correlated ucl algorithm for the correlated mab problem . we first propose a modified ucl algorithm , and then analyze its performance .
the modification is designed to leverage prior information on correlation structure .
suppose the prior on the mean rewards at each arm is a multivariate gaussian random variable with mean vector @xmath64 and covariance matrix @xmath65 . for the above mab problem ,
the posterior distribution of the mean rewards at each arm at time @xmath12 is a gaussian distribution with mean @xmath66 and covariance @xmath67 defined by @xmath68 where @xmath69 is the column @xmath0-vector with @xmath2-th entry equal to one , and every other entry zero . in the following ,
we denote entries of @xmath70 and the diagonal entries of @xmath67 by @xmath71 and @xmath72 , respectively . as in section [ subsec : deterministic - ucl ] , let @xmath28 be the number of times arm @xmath15 has been selected until time @xmath12 , and @xmath29 be the empirical mean of the rewards from arm @xmath15 until time @xmath12 .
then , it is easy to verify that @xmath73 where @xmath74 , @xmath75 is the diagonal matrix with entries @xmath76 , and @xmath77 is the vector of @xmath78 .
the correlated ucl algorithm for the gaussian mab problem , at each decision instance @xmath3 , selects an arm with the maximum upper credible limit , i.e. , it selects an arm @xmath34 , where @xmath79 @xmath36 is the inverse cumulative distribution function for the standard gaussian random variable , @xmath37 , @xmath80 is the correlation coefficient between arm @xmath15 and arm @xmath81 at time @xmath12 and @xmath38 and @xmath39 are tunable parameters .
note that for uncorrelated priors , @xmath82 and the correlated ucl algorithm reduces to the ucl algorithm . in the context of uninformative priors , @xmath83 for each @xmath55 , and
the ucl algorithm selects each arm once in first @xmath0 steps . in a similar vein
, we introduce an initialization phase for the correlated ucl algorithm . _
* initialization : * _ in the initialization phase , an arm @xmath2 defined by @xmath84 is selected at time @xmath12 . here
, @xmath85 is a pre - specified positive constant .
let @xmath86 be the number of steps in the initialization phase .
[ lem : initialization - corr - stat ] for the correlated mab problem and the inference process , the initialization phase ends in at most @xmath0 steps and the variance following the initialization phase @xmath87 , for each @xmath5 .
note that to prove the lemma , it suffices to show that no arm will be selected twice in the initialization phase .
it follows from the sherman - morrison formula for the rank-@xmath88 update for the covariance in that @xmath89 where @xmath90 is the @xmath91 component of @xmath67 , for each @xmath5 . if @xmath92 , then @xmath93 .
thus , arm @xmath81 will not be selected again in the initialization phase which establishes our claim .
[ rem : corr - init ] for correlated priors , the inference equations yield the following expressions for the bias @xmath94 and covariance @xmath95 of the estimate @xmath70 @xmath96 - \bs m = ( \lambda_0 + p(t)^{-1})^{-1 } \lambda_0 ( \bs \mu_0 - \bs m)\\ \bar \sigma(t ) & : = \text{cov}(\bs \mu_t ) = ( \lambda_0 + p(t)^{-1})^{-1 } p(t)^{-1}(\lambda_0 + p(t)^{-1})^{-1},\end{aligned}\ ] ] where @xmath97 is the vector of mean reward .
let @xmath98 and @xmath99 , @xmath100 be the diagonal and off - diagonal entries of @xmath67 , and @xmath101 be the diagonal entries of @xmath102 .
we now analyze the properties of covariance matrices @xmath67 and @xmath102 .
let @xmath103 be the submatrix of @xmath104 obtained after excluding the @xmath15-th row and @xmath15-th column .
let @xmath105 be the row vector obtained after excluding the @xmath15-th entry from the @xmath15-th row of @xmath104 .
we define the variance of arm @xmath15 conditioned on the mean reward at every other arm by @xmath106 let @xmath107 . with a slight abuse of notation , we refer to @xmath28 as the number of times arm @xmath15 is selected after the initialization phase .
we also define for each @xmath5 @xmath108 where @xmath109 is the @xmath110 component of @xmath111 .
[ lem : bounds - variances ] the following statements hold for the inference process : 1 .
the variance @xmath98 satisfies @xmath112 2 . the variance @xmath113 satisfies @xmath114 we start by establishing the first statement .
the covariance update in can be simplified using the sherman - morrison formula to obtain @xmath115 it follows that @xmath116 it follows that after the initialization phase @xmath117 . moreover , at each future round , if @xmath118 , then @xmath119 ; otherwise , @xmath120 .
the upper bound on @xmath98 immediately follows from this observation and the induction argument .
we now establish the lower bound on @xmath98 .
since the inference process involves a stationary environment , the sequence in which arms are played is of no significance and the inference only depends on the number of times an arm has been played . consequently , the inference is the same if arms are played in blocks . in particular ,
each arm @xmath121 can be played in a block of size @xmath122 .
further , any order in which these blocks are played leads to the same inference .
suppose for such a modified allocation of arms , @xmath123 is the time when the block associated with arm @xmath81 begins .
suppose that arm @xmath15 is played the last .
then , from and for the modified allocation process , it follows that @xmath124 i.e. , the posterior variance @xmath125 is lower bounded by the conditional variance of arm @xmath15 under a noise free reward from arm @xmath81 .
it follows that , for the modified allocation sequence , @xmath126 .
now , the lower bound follows from the variance update after the last block . to establish the second statement ,
we note that @xmath127 .
it follows that @xmath128 where the second inequality follows from the fact @xmath129 .
similarly , @xmath130 establishing the lower bound .
[ thm : corr - regret ] for the gaussian mab problem , and the correlated ucl algorithm , the expected number of times a suboptimal arm @xmath15 is selected after the initialization phase satisfies @xmath60 \le \eta_i + \hat n_i(t),\ ] ] where @xmath131 , and @xmath132 see [ proof : corr - regret ] .
[ rem : corr - regret ]
in this section , we illustrate the results of the preceding two sections with data from numerical simulations .
the theoretical results pertain to different quality priors defined by how rich is the information they can capture about the rewards associated with the bandit .
uninformative priors capture no information , while uncorrelated informative priors capture beliefs about individual arms .
correlated ( informative ) priors add to uncorrelated informative priors the ability to capture beliefs about the relationship between different arms , which we leverage in our new correlated ucl algorithm .
when an informative prior models the environment well , we refer to it as a _ well - informed _ prior ; conversely , if the prior models the environment poorly , we refer to it as _ ill - informed_. as in @xcite , our simulations focus on the case of a spatially - embedded bandit problem , for which @xcite showed that correlated priors can lead to higher performance .
the simulations show that , among well - informed priors , those with richer information content result in higher performance .
theorems [ thm : uncorr - regret ] and [ thm : corr - regret ] allow us to quantify the extent to which a prior is well - informed .
we consider here the spatially - embedded bandit problem studied in @xcite .
the reward surface is relatively smooth with regions of both high and low rewards .
this means that a correlated prior capturing length scale information can improve performance .
the mean reward value is equal to 30 , and the sampling variance for each arm is @xmath133 .
figure [ fig : goodpriors ] shows simulations from cases where the informative priors are well - informed .
mean cumulative regret computed from an ensemble of 100 simulations is shown for three priors : an uninformative prior , an informative uncorrelated prior , and an informative correlated prior . for all the simulations , the parameter @xmath134 was set equal to @xmath135 , and for correlated priors the parameter @xmath136 was set equal to 1 .
the informative priors have an initial mean belief @xmath137 with a higher value ( equal to 100 ) in regions with high rewards , and a lower value of zero elsewhere .
the uncorrelated prior sets @xmath138 , meaning the prior represents the equivalent of a single prior observation .
the correlated prior sets @xmath139 as in the uncorrelated case , and uses a correlation structure representing an exponential kernel as in @xcite .
this kernel encodes the information that the closer two arms are in the embedding space , the more correlated are their rewards . the richer information provided by the informative priors results in better performance in this case where the priors are well - informed : the informative correlated prior results in less regret than the informative uncorrelated prior , which in turn results in less regret than the uninformative prior . for short horizons
, the informative priors result in cumulative regret which is less than the lai - robbins lower bound . the ucl algorithm and
the correlated ucl algorithm can violate the lower bound because of the additional information provided by the priors , which effectively shifts the regret curve leftwards .
asymptotically , however , the algorithms will tend to match the lai - robbins regret rate for any prior .
in contrast , figure [ fig : badpriors ] shows simulations from cases where the informative priors are variously ill - informed .
mean cumulative regret computed from an ensemble of @xmath140 simulations is shown for three increasingly informative priors , as in figure [ fig : goodpriors ] .
the informative priors have an initial mean belief @xmath137 that is uniform with each element @xmath141 . as in figure
[ fig : goodpriors ] , the uncorrelated prior sets @xmath138 , meaning the prior represents the equivalent of a single prior observation .
the correlated prior sets @xmath139 and uses a correlation structure that again represents an exponential kernel but with a longer length scale to represent a smoother reward surface .
although the informative priors accurately represent the overall mean value of the reward surface , they fail to capture the spatial heterogeneity of the reward surface , in particular the fact that it has high- and low - value patches .
therefore , both informative priors are ill - informed about the mean rewards and the informative uncorrelated prior results in much poorer performance than the uninformative prior for moderate task horizons . however ,
by adding the correlation structure to the ill - informed uncorrelated prior , we can recover much of the performance exhibited by the well - informed correlated prior of figure [ fig : goodpriors ] .
in a spatially - embedded task like the one studied here , information about correlation structure among arms can be as valuable as accurate information about the value of individual arms . .
because of the additional information provided by the informative priors , the algorithms can sample arms more selectively from the initial time @xmath142 , which results in better performance than the uninformative prior and allows the algorithms to outperform the lai - robbins bound on regret.,scaledwidth=50.0% ] , the traces show mean cumulative regret from 100 simulations for each of three different priors .
again the algorithms exhibit an initialization phase behavior for the uninformative and informative correlated priors , whose end can be seen in the bends in the regret curves near @xmath143 .
the ill - informed correlated prior improves performance relative to the uninformative prior although not quite as much as the well - informed correlated prior does in figure [ fig : goodpriors ] .
in contrast , the ill - informed uncorrelated prior significantly decreases performance relative to all other priors . by encoding a strong incorrect belief about the rewards , this
prior requires multiple samples of suboptimal arms to learn that they are suboptimal .
this appears in the regret curve as an initialization phase that lasts until @xmath144 4,500 , at which point the mean cumulative regret is approximately 35,000.,scaledwidth=50.0% ]
in this note we studied and modified the ucl algorithm for the correlated mab problem with gaussian rewards .
we investigated the influence of the assumptions in the prior on the performance of the ucl algorithm and the new correlated ucl algorithm .
we characterized scenarios in which the informative priors perform better than the uninformative prior and characterized the improvement in the performance in terms of cumulative regret .
in particular , we showed conditions in which an informative correlated prior can be leveraged to significantly reduce cumulative regret .
there are several possible avenues of future research .
first , we considered that the environment is stationary .
an interesting future direction is to consider non - stationary environments in which the reward at each arm may be time - varying and the autocorrelation scale may be known .
second , we considered these problems for a single player .
many application scenarios involve a group of individuals and it is of interest to study collaborative and competitive multiplayer versions of these problems .
10 j. gittins , k. glazebrook , and r. weber . .
wiley , second edition , 2011 .
e. kaufmann , o. capp , and a. garivier .
on bayesian upper confidence bounds for bandit problems . in _ international conference on artificial intelligence and statistics _ , pages 592600 , la palma , canary islands , spain , april 2012 .
n. srinivas , a. krause , s. m. kakade , and m. seeger .
information - theoretic regret bounds for gaussian process optimization in the bandit setting .
, 58(5):32503265 , 2012 .
s. agrawal and n. goyal .
analysis of thompson sampling for the multi - armed bandit problem . in s.
mannor , n. srebro , and r. c. williamson , editors , _ jmlr : workshop and conference proceedings _ , volume 23 : colt 2012 , pages 39.139.26 , 2012 .
p. reverdy , v. srivastava , and n. e. leonard .
modeling human decision making in generalized gaussian multiarmed bandits .
, 102(4):544571 , 2014 .
h. robbins . some aspects of the sequential design of experiments .
, 58:527535 , 1952 .
r. agrawal , m. v. hedge , and d. teneketzis .
asymptotically efficient adaptive allocation rules for the multi - armed bandit problem with switching cost .
, 33(10):899906 , 1988 .
v. anantharam , p. varaiya , and j. walrand . .
, 32(11):968976 , 1987 .
j. l. ny , m. dahleh , and e. feron .
multi - uav dynamic routing with partial observations using restless bandit allocation indices . in _ proceedings of american controls
conference _ , pages 42204225 , seattle , washington , usa , june 2008 .
m. y. cheung , j. leighton , and f. s. hover .
autonomous mobile acoustic relay positioning as a multi - armed bandit with switching costs . in _
ieee / rsj international conference on intelligent robots and systems _ , pages 33683373 , tokyo , japan , november 2013 .
v. srivastava , p. reverdy , and n. e. leonard .
surveillance in an abruptly changing world via multiarmed bandits . in _
ieee conf . on decision and control _ ,
pages 692697 , los angeles , ca , december 2014 .
m. babaioff , y. sharma , and a. slivkins .
characterizing truthful multi - armed bandit mechanisms . in _ proceedings of the 10th acm conference on electronic commerce _ , pages 7988 , stanford , ca ,
usa , july 2009 .
f. radlinski , r. kleinberg , and t. joachims . learning diverse rankings with multi - armed bandits . in _ proceedings of the 25th international conference on machine learning _ , pages 784791 , helsinki , finland , july 2008 .
mccall and j. j. mccall . a sequential study of migration and job search .
, 5(4):452476 , 1987 .
j. r. krebs , a. kacelnik , and p. taylor .
test of optimal sampling by foraging great tits .
, 275(5675):2731 , 1978 .
v. srivastava , p. reverdy , and n. e. leonard .
optimal foraging and multi - armed bandits . in _
allerton conf . on communications , control and computing _ ,
pages 494499 , monticello , il , usa , october 2013 .
r. c. wilson , a. geana , j. m. white , e. a. ludvig , and j. d. cohen .
humans use directed and random exploration to solve the explore - exploit dilemma .
, 143(6):20742081 , 2014 .
p. reverdy , v. srivastava , r. c. wilson , and n. e. leonard .
human decision making and the explore - exploit tradeoff : algorithmic models for multi - armed bandit problems . in simon haykin ,
editor , _ cognitive dynamic systems_. wiley series on adaptive and cognitive dynamic systems .
ieee press / wiley , 2015 .
t. l. lai and h. robbins .
asymptotically efficient adaptive allocation rules . , 6(1):422 , 1985 .
p. auer , n. cesa - bianchi , and p. fischer .
finite - time analysis of the multiarmed bandit problem .
, 47(2):235256 , 2002 .
s. bubeck and n. cesa - bianchi .
regret analysis of stochastic and nonstochastic multi - armed bandit problems .
, 5(1):1122 , 2012 .
e. kaufmann , n. korda , and r. munos .
thompson sampling : an asymptotically optimal finite - time analysis . in _
algorithmic learning theory _ ,
pages 199213 .
springer , 2012 . w. r. thompson . on the likelihood that one unknown probability exceeds another in view of the evidence of two samples .
, 25(3/4):285294 , 1933 . c. y. liu and l. li . on the prior sensitivity of thompson sampling
. , 2015 .
m. abramowitz and i. a. stegun , editors . .
dover publications , 1964 .
in the spirit of @xcite , we bound @xmath14 as follows : @xmath145 where @xmath146 is some positive integer and @xmath147 is the indicator function , with @xmath148 if @xmath149 is a true statement and @xmath150 otherwise . at time @xmath12 , the agent picks option @xmath15 over @xmath18 only if @xmath151 this is true when at least one of the following equations holds : @xmath152 where @xmath153 and @xmath37 .
otherwise , if none of the equations ( [ eq : qi*])-([eq : micomp ] ) holds , @xmath154 and option @xmath18 is picked over option @xmath15 at time t. as noted earlier , the posterior mean @xmath71 is a gaussian random variable : @xmath155 we will now analyze the events , , and . let @xmath156 be the probability of the event . [ lem - first - event - uncorr ] the following statements hold for event : 1 .
if @xmath157 , then @xmath158 2 . if @xmath159 , then @xmath160 for @xmath161 , event is true if @xmath162 where @xmath163 is a standard normal random variable .
similarly , for @xmath164 , event is not true if ( i ) @xmath165 , or ( ii ) @xmath166 and @xmath167 .
we now establish the first statement . if @xmath165 and @xmath168 , then @xmath169 . if @xmath165 and @xmath170 , then @xmath171 therefore , @xmath172 to establish the second statement , we note that if @xmath173 and @xmath168 , then event does not hold if @xmath174 if @xmath173 and @xmath170 , then @xmath175 , where @xmath176 .
note that @xmath177 , if @xmath178 .
define @xmath179 it follows that for @xmath180 , @xmath181 where the second last inequality follows from lemma [ lem : diff - of - squares ] and @xmath182 and @xmath183 are as defined in section [ sec : uncorr - ucl - regret ] . therefore , @xmath184 let @xmath185 be the joint probability of the event and the event @xmath186 , for some @xmath187 .
[ lem - sec - event - uncorr ] the following statements hold for event : 1 .
if @xmath188 , then @xmath189 2 . if @xmath190 , then @xmath191 the event ( [ eq : qi ] ) holds if @xmath192 where @xmath163 is a standard normal random variable .
we start with establishing the first statement . if @xmath193 and @xmath194 , then @xmath195 where @xmath196 .
it follows that @xmath177 , if @xmath197 .
it follows that for @xmath198 @xmath199 where the second last inequality follows from lemma [ lem : diff - of - squares ] .
therefore , @xmath200 the second statement follows similarly to the first statement in lemma [ lem - first - event - uncorr ] .
we now analyze the probability of event .
@xmath201 where @xmath17 , the inequality follows from lemma [ lem : ineq ] , and the inequality follows from the monotonicity of the logarithmic function .
therefore , the event ( [ eq : micomp ] ) is not true if @xmath202 setting @xmath61 , we get @xmath203 } } & \leq \eta_i + \sum_{t = 1}^t \prob(q_{i}^t > q_{i^*}^t , n_{i}(t-1 ) \geq \eta_i ) \\ & = \eta_i + \sum_{t = 1}^t \big(\prob_1(t ) + \prob_2(t)\big)\\ & <
\eta_i + \hat n_i(t ) . \end{aligned}\ ] ] this completes the proof of the theorem .
similar to the proof of theorem [ thm : uncorr - regret ] , at time @xmath12 , the agent picks option @xmath15 over @xmath18 only if @xmath204 .
this is true when at least one of the following equations holds : @xmath205 where @xmath206 , @xmath37 .
it follows using the same argument as in theorem [ thm : uncorr - regret ] that @xmath214 similarly , @xmath215 also , event is not true if @xmath216 adding the probabilities of the events - , we obtain the desired expression . | we consider the correlated multiarmed bandit ( mab ) problem in which the rewards associated with each arm are modeled by a multivariate gaussian random variable , and we investigate the influence of the assumptions in the bayesian prior on the performance of the upper credible limit ( ucl ) algorithm and a new correlated ucl algorithm .
we rigorously characterize the influence of accuracy , confidence , and correlation scale in the prior on the decision - making performance of the algorithms .
our results show how priors and correlation structure can be leveraged to improve performance . multiarmed bandit problem , bayesian algorithms , decision - making , spatial search , upper credible limit algorithm , influence of priors |
since the early years of superconductivity studies it has been noticed that any superconducting transition @xmath0 always has a finite width .
very often sample inhomogeneity is the dominating factor .
however with refinement of fabrication methodology it became clear that even in the most homogeneous samples the @xmath0 transition is not infinitely narrow .
the broadening of the @xmath0 dependence is most pronounced in systems with reduced dimensions where the thermodynamic fluctuations have a larger effect . in particular case of quasi - one - dimensional ( 1d ) channels fluctuation - driven phase slips ,
the concept first introduced by w. little in 1967 @xcite , are responsible for the finite resistivity in a narrow region below the critical temperature @xcite @xmath1 , where @xmath2 is the condensation energy of the smallest statistically independent volume of the wire . soon after formulation of the model @xcite experiments
@xcite confirmed the validity of the concept of the thermally activated phase slips ( taps ) . at lower temperatures the number of thermally activated phase slips
is exponentially suppressed and no resistance should be expected well below the critical temperature .
however later experiments @xcite in extremely narrow superconducting nanowires have demonstrated the finite resistance even at temperatures @xmath3 .
the effect has been associated with quantum fluctuations of the order parameter , alternatively called _ quantum phase slips _ ( qps ) @xcite .
yet another confirmation of the qps reality came from the experiments studying the persistent current in nanorings @xcite resulting in building a quantum two level system - qubit @xcite .
though the quantum fluctuation phenomenon has already received the experimental confirmation from several independent sources , the physics behind is still under debates .
of special interest are the recent theoretical developments predicting that a qps junction is dual to a josephson junction @xcite .
contrary to conventional transport measurements with low - ohmic contacts @xcite , if a nanowire governed by qpss ( qps junction ) is truly current biased , then one should expect the development of the insulating state - the coulomb blockade @xcite in a full accordance with a josephson junction in the regime of bloch oscillations @xcite the mandatory pre - requisites of such a non - trivial experimental configuration is the high enough rate of qpss being of the same order as the experimentally observed coulomb gap .
contrary to such a sample current biased through high - ohmic electrodes , similar superconducting nanowire of the same diameter , but with low - ohmic probes ( e.g. superconducting ) , should not demonstrate the coulomb gap , but its @xmath0 and @xmath4 dependencies should show no traces of conventional superconductivity either .
the objective of this paper is to study exactly the limit of extremely narrow nanowires where the qps rate is high enough to significantly affect the transport properties .
the impact of the quantum fluctuations on the shape of the superconducting transition is qualitatively described by expression , similar to the taps expression , @xcite @xmath5 , where instead of the thermal energy @xmath6 stands the rate of qps @xmath7 where @xmath8 is the normal state resistance , @xmath9 is the superconducting quantum resistance , @xmath10 is the length of the wire , @xmath11 and @xmath12 are the temperature - dependent superconducting energy gap and coherence length .
@xmath13 is the qps action , where constant @xmath14 is of the order of unit and can not be more precisely defined from the theory @xcite .
it is the only true fitting parameter of the model , other parameters can be derived from the experimental data . for our dirty limit samples @xmath15 the superconducting coherence length can be estimated as @xmath16 , where @xmath17 is the coherence length in the clean limit @xmath18 and @xmath19 is the fermi velocity . utilizing the text - book expression for the normal state resistance of a wire with cross - section @xmath20 , length @xmath10 and resistivity @xmath21
, one comes to a conclusion that the effective resistance of a qps - governed system exponentially depends on the sample parameters @xmath22 .
hence thin nanowires of low-@xmath23 superconductors with high resistivity ( in normal state ) are of advantage for observation of a pronounced contribution of the qps effect .
it should be noted that the model @xcite is based on the assumption that quantum fluctuations are relatively rare events . or in other terms , the corresponding qps - related effective resistance of the nanowire should be much smaller than the normal state resistance @xmath24 .
as it comes from the formulated above objective of the paper , the limit of strong fluctuations violates this requirement .
unfortunately , to our best knowledge , the limit of strong quantum fluctuations has not yet been properly treated theoretically .
hence , the expression ( [ eq1 ] ) can be considered only as a certain guideline . as it will be shown below , indeed when
the contribution of the qps effect on the shape of the @xmath0 transition is already noticeable , the further reduction of the nanowire cross - section @xmath20 leads to complete flattening of the @xmath0 dependence not described by the model @xcite .
the research related to mesoscopic superconducting is limited by the fabrication technology . in particular case of qps effect , to reach the regime of interest @xmath25
the nanowire should have very small cross - section .
for example for aluminum the cross - section below 10 nm ( still maintaining the high level of uniformity ! ) is mandatory @xcite .
following eq .
( 1 ) , for materials like nb , sn or pb with relatively high critical temperature one gets even more pessimistic estimations approaching 1 nm scale .
this note explains why in particular case of niobium no traces of the qps effect have been observed down to 7 nm scales @xcite . with proper material selection
the extremely tough fabrication requirements can be somehow relaxed . from our previous studies
we have already learnt that in titanium the qps effect is observable at sub-40 nm scales @xcite , which is achievable with the standard e - beam lithography technique .
in addition , titanium is an easy to work material and the extended microscopic and elemental analysis reveals no severe structural defects or / and impurities @xcite . for the purpose of the present work we prepared several titanium nanowires with length @xmath26 and the effective diameter @xmath27 between 27 and 48 nm .
the structures were fabricated with the standard lift - off e - beam lithography , and the titanium was deposited in uhv e - beam evaporator at residual pressure @xmath28 mbar . the substrate covered with exposed pmma
/ maa mask was cleaned with low - energy @xmath29 plasma immediately before the metal evaporation .
based on the measured resistivity @xmath30 @xmath31 , which is comparable with the tabulated value for the clean bulk titanium , the quality of the metal thin film is acceptable .
previous tem and tof - erda analyses of titanium nanostructures fabricated using identical conditions and parameters confirmed the high material quality : the highest concentration of foreign inclusions inside the titanium matrix corresponds to @xmath32 @xmath33 of oxygen @xcite .
this low concentration of impurities , combined with tem analysis , disables any speculations about the presence of sample inhomogeneities capable to significantly broaden the @xmath0 dependence .
the spm analysis of the studied samples ( figure 1 ) confirmed the standard sample quality with the surface roughness of about several nm @xcite . the mean free path @xmath34 for the wire can be evaluated from the measured normal state resistance @xmath8 and the known material constant @xmath35 varying slightly in different literature sources . for our samples
estimations give the mean free path @xmath36 1 to 2 nm .
the result agrees well with the spm and tem measured grain size of roughly 2 to 3 nm .
substrate.,width=302 ] the @xmath0 measurements ( figure 2 ) were made employing the conventional four - probe configuration using 7 hz lock - in technique at rather small excitation currents of about 50 pa .
the experiments were made in @xmath37 dilution refrigerator with the base temperature down to 17 mk .
all input / output lines were carefully protected from the noisy electromagnetic environment utilizing multi - stage rlc filtering enabling the electron temperature @xmath3835 mk at a base ( phonon ) temperature @xmath3820 mk @xcite .
for sufficiently thick nanowires with the effective diameter @xmath39 40 nm the shape of the @xmath0 dependencies follows our earlier findings : pronounced superconducting transition with the low temperature data @xmath40 which can be described by the qps model @xcite with the realistic set of fitting parameters @xcite . however the thinner the sample , the less the resistance @xmath0 drops below the certain critical temperature ( figure 2 ) .
it should be noted that for titanium the critical temperature @xmath23 decreases with the decrease of the cross - section @xmath20 .
the size dependence of the critical temperature for low dimensional superconductors is a well - known effect , though the commonly accepted explanation of the phenomenon still does not exist @xcite .
thus the accurate experimental definition of the critical temperature for the thinnest samples @xmath41 , necessary for the theory fitting ( equation [ eq1 ] ) , is rather problematic due to the strongly broadened @xmath0 transition .
nevertheless for the thinnest samples @xmath42 35 nm no set of realistic qps model parameters can fit the experimental data ( figure 2 ) . for these ultra - thin samples
the experimental @xmath0 dependency is so weak , that the model @xcite applicability criterion @xmath43 is not satisfied down to the lowest experimentally obtainable temperatures .
and the effective diameters indicated in the inset .
model @xcite calculations using the realistic fitting parameters @xmath44 1 nm , @xmath45=1.79@xmath46 m / s , @xmath14=0.098 and the critical temperature @xmath23=200 mk and 115 mk for the 37 nm and 32 nm samples , respectively , are plotted with dashed lines .
best fit @xmath23 s are indicated with arrows .
clearly the model @xcite , applicable in the limit @xmath43 , fails to provide any acceptable fits for the thinnest samples @xmath0 data.,width=302 ] at temperatures well - below the critical temperature @xmath3 the @xmath47 dependencies ( figure 3 ) for the thickest studied samples demonstrate the conventional destruction of the zero - resistance state by current with the well - pronounced critical current @xmath48 .
the double shape of the transition ( e.g. at @xmath3822 na and @xmath3828 na for the 48 nm sample ) presumably originates from slightly different effective cross - section of the samples at the node regions . with decrease of the nanowire cross - section ,
as expected from the @xmath0 dependencies , the the zero - resistance state disappears and only some residual critical current peculiarities can be traced . in the thinnest samples the zero - biased differential resistance @xmath49 very slightly increases indicating the presence of a weak coulomb blockade .
observation of a pronounced coulomb effects requires the true current biasing ( including the high frequencies ! ) @xcite and this experimental realization is not within the scope of the present paper .
, as function of the bias current i for the same four samples as in figure 2 .
the data for the 48 nm nanowire ( open right - pointing triangles ) have been collected at 80 mk , while for the other three samples - at 20 mk . for clarity
the curves are vertically shifted .
the scale corresponding the normal state resistance @xmath8 is indicated with vertical bar.,width=302 ] yet one more interesting observation is the negative magnetoresistance ( nmr ) which is observed in the thinnest samples with pronounced qps contribution ( figure 4 ) .
the effect increases with lowering the temperature .
similar phenomenon has been reported in aluminum @xcite , niobium and @xmath50 @xcite and lead @xcite nanowires . the origin of the nmr in these quasi-1d channels is still under debates . in case of niobium and @xmath50
it was conjected that some rogue magnetic moments might be present , and their pair breaking contribution , active at lower magnetic fields , is suppressed by higher fields leading to the observed nmr @xcite .
however this mechanism needs an independent proof of the existence of these magnetic moments , and can not explain the nmr in such material as aluminum @xcite , where it is a well - known fact that majority of magnetic impurities can obtain a non - negligible magnetic moment only at relatively high concentrations @xcite .
another explanation of the nmr phenomenon deals with the magnetic field suppression of the charge imbalance accompanying each phase slip event @xcite .
somehow related mechanism , capable to provide nmr , deals with more effective suppression of superconductivity in ( wider ) electrodes affecting the phase slip formation in the ( thinner ) nanowire @xcite .
measured at zero magnetic field ( filled triangles ) and in perpendicular magnetic field b = 5.6 mt ( open circles ) .
both data were obtained while the same slow temperature sweep just by switching on and off the magnetic filed .
the inset shows the representative region with the negative magnetoresistance , measured at a constant temperature t = 100 mk .
, width=302 ]
@xmath51 and @xmath52 characteristics of titanium nanowires with sub-50 nm diameters were measured .
the shape of the @xmath0 dependencies for the not - too - narrow samples with the effective diameters @xmath39 40 nm confirms the earlier findings : pronounced superconducting transition with the low temperature data @xmath40 which can be described by the quantum phase slip model @xcite . in thinner samples
the @xmath0 transitions dramatically flatten disabling any comparison with the existing fluctuation models , which assume that the phase slips are still the rare events and hence the effective resistance should be much smaller than the normal state resistance .
the @xmath47 dependencies confirm the @xmath0 data conclusion about the absence of the truly zero - resistance state in the thinnest samples .
the negative magnetoresistance is observed , while the origin of the effect is not clear .
the national graduate school in materials physics ( ngsmp , finland ) is gratefully acknowledged for financial support . | in a number of recent experiments it has been demonstrated that in ultra - narrow superconducting channels quantum fluctuations of the order parameter , alternatively called quantum phase slips , are responsible for the finite resistance well below the critical temperature .
the acceptable agreement between those experiments and the models describing quantum fluctuations in quasi - one - dimensional superconductors has been established .
however the very concept of the phase slip is justified when these fluctuations are the relatively rare events , meaning that the effective resistance of the system should be much smaller than the normal state equivalent . in this paper
we study the limit of the strong quantum fluctuations where the existing models are not applicable .
in particular case of ultra - thin titanium nanowires it is demonstrated that below the expected critical temperature the resistance does not demonstrate any trend towards the conventional for a superconductor zero - resistivity state even at negligibly small measuring currents .
application of a small magnetic field leads to an unusual negative magnetoresistance , which becomes more pronounced at lower temperatures .
the origin of the negative magnetoresistance effect is not clear . |
99 f. nozzoli , presentation given at this conference ( taup 2009 ) http://taup2009.lngs.infn.it/slides/jul3/nozzoli.pdf or http://people.roma2.infn.it/ dama / pdf / nozzoli_taup09.pdf .
r. bernabei et al .
instr . & meth . a 592 ( 2008 ) 297 .
r. bernabei et al .
, isbn 978 - 88 - 95688 - 12 - 1 , pages 1 - 53 ( 2009 ) exorma ed .
( arxiv:0806.0011[astro - ph ] ) .
see e.g. r.m .
kippen , new astr .
48 ( 2004 ) 221 .
v. kundryavtsev , presentation given at this conference ( taup 2009 ) ; v. kudryavtsev et al . ,
r. bernabei et al .
j. c 56 ( 2008 ) 333 .
r. bernabei et al .
phys . a 21 ( 2006 ) 1445 .
r. bernabei et al .
d 77 ( 2008 ) 023506 .
r. bernabei et al . , mod .
lett . a 23 ( 2008 )
r. bernabei et al . , int . j. mod
a 22 ( 2007 ) 3155 .
r. bernabei et al .
j. c 53 ( 2008 ) 205 .
j. angle et al .
( 2008 ) 021303 .
a. benoit et al . , phys .
b 637 ( 2006 ) 156 . | a variety of detectors has been proposed for dark matter direct detection , but most of them by the fact are still at r&d stage . in many cases , it is claimed that the lack of an adequate detectors radio - purity might be compensated through heavy uses of montecarlo simulations , subtractions and handlings of the measured counting rates , in order to claim higher sensitivity ( just for a particular scenario ) .
the relevance of a correct evaluation of systematic effects in the use of montecarlo simulations at very low energy ( which has always been safely discouraged in the field so far ) and of multiple subtractions and handling procedures applied to the measured counting rate is shortly addressed here at some extent .
many other aspects would also deserve suitably deep investigations . in this paper
some arguments presented at the taup09 conference will be shortly summarized .
more details , tables and figures can be found in the slides at the conference site@xcite .
let us firstly comment the possibility of reliable evaluations of the background contributions at the kev energy region in the field of dark matter searches . as well known , it has been generally discouraged this procedure in the field of dark matter over more than twenty years .
in fact , the estimation by a montecarlo simulation of the background component in the counting rate from the residual radioactivity requires a detailed knowledge of : i ) the exact set - up geometry ( detector or detectors matrix , all materials , details of the assembling , of the shield layers , of the site , etc . ) ; ii ) the detector response function ( e.g. energy resolution , @xmath0 ratio , channeling , etc . ) ; iii ) the nature , the position and the concentration of all the existing radioactive contaminants ; iv ) etc .. unfortunately , apart from the geometrical layout of the set - up that are generally well known by people inside the experimental group , all the other quantities necessary in the montecarlo simulation require dedicated measurements .
moreover , there are some quantities ( such as concentration of residual contaminants , etc . ) that can be poorly known and just upper / lower limits are available ; in some cases these quantities can be even totally unknown .
as an example , the experimental energy resolution as a function of the energy and the energy scale should be measured / verified down to the energy threshold , as done e.g. by the dama / libra experiment where they are continuously measured by external / internal known sources from mev down to the energy threshold @xcite . on the contrary , in other experiments these quantities
are instead extrapolated from calibrations at much higher energy ( as done e.g. by liquid noble gas set - ups , where the energy threshold and the few kev energy scale are generally unproven , also because of position dependence , of non - uniform signal collection , etc . ; see e.g. @xcite ) .
regarding the presence of residual contaminants in the set - up , generally only limits on the contributions of the `` standard '' contaminants are given ; these limits forbid any reliable estimation of the background ( being unknown the exact values ) and can not be obviously exceeded ( see also later ) .
moreover , possible presence of many non - standard contaminants should be also included .
in addition , the montecarlo simulation also depends on the precise location of all the contaminants that is generally unknown even for the `` standard '' ones in complex set - ups .
the situation is more complex for multi - detectors set - up and when the energy distribution refers to events where each detector has all the others in anticoincidence ( _ single - hit _ events ) .
thus , it is trivial to conclude that a reliable precise simulation of the background counting rate in particular at kev energy region is not univocally determined and is a quite impossible task .
in addition , beyond the fore - mentioned arguments , we need to take into account that a montecarlo code can not manage all the possible low energy atomic physical processes .
this argument is still subject of improvements ; in fact , as an example , non - negligible differences are also obtained by different versions of the same montecarlo code @xcite .
some instructive examples are given by the trails in montecarlo simulations in ref .
@xcite .
as shown there , these simulations noticeably differ from the measured energy distributions in the cases of xenon-10 and of zeplin - iii .
in fact , they predict twice the measured rate for xenon-10 near 200 kev , and more than one order of magnitude the measured rate for zeplin - iii in the mev range @xcite .
the same approach has been also pursued by the same authors @xcite trying a montecarlo simulation of the background in the dama / libra set - up .
in particular , apart from errors in the details of the set - up geometry reconstruction and in the multiple - hit definition , many crude and arbitrary approximations in the nature and in the location of the residual contaminants have been arbitrarily assumed ; in fact , e.g. : i ) only standard contaminants , ii ) only unbroken chains , iii ) only uniform location of contaminants in the detectors , etc . have been taken into account . as a result of this rough , partial and arbitrary approach ,
the predicted rate has been estimated within a factor 10 lower than the measured one . instead of refining the quality of the simulation or reasonably recognizing the impossibility of precise determination , the authors just pursued the exercise of arbitrarily increasing `` by hand '' the assumed values of the contaminants at levels much larger than the measured experimental limits @xcite .
this also implies an overestimate of the background in higher energy region with respect to the measured experimental rate . in conclusion , although the arbitrary and the erroneous adopted procedures , these authors do not succeed in reproducing either the low or the high energy spectra .
nevertheless , as a conclusion of this arbitrary exercise @xcite , this artificially - boosted simulated spectrum has been subtracted by the measured one , attempting to obtain a limit for the unmodulated dark matter signal component .
this example shows how subtraction procedures using montecarlo simulations in the few kev energy region can give rise to erroneous conclusions ; thus , any constraint on dark matter signal on this basis would be an artefact .
furthermore , let us also note that the measured spectra e.g. of the existing / past nai(tl ) detectors ( such as e.g. anais , frejus , naiad , elegant , etc ... ) do not support even the shape presented in ref .
@xcite .
in addition , well different counting rates at kev energy region are present even for detectors of the same experimental group , as e.g. the case of naiad in 1996 and in 2003 @xcite . in conclusion
, it does not exist an unique recipe for a precise and reliable montecarlo simulation of whatever set - up , and for nai(tl ) in particular .
let us finally remind that a safer approach has been presented in this conference by dama collaboration ; this shows that enough space is present in the measured counting rate of the dama / libra kev energy spectrum for the unmodulated component of dark matter signal @xcite .
in the second part of this contribution @xcite , problems related to the application of multiple subtraction procedures of the measured counting rate , as pursued by experiments trying to identify the presence of recoil nuclei in the measured energy spectrum , have been summarized . in fact , many of the existing dark matter candidates also in the wimp class can give rise to signals that either have totally an electromagnetic nature ( see e.g. @xcite ) or involve electromagnetic signals associated to nuclear recoils ( see e.g. @xcite ) ; obviously , approaches that are based on multiple subtraction procedure of the electromagnetic component of the counting rate are blind to similar scenarios .
moreover , well known side processes exist for recoils ( such as recoils induced by neutrons , fission fragments , end - range alphas , surface electrons , etc . ) .
this approach is generally pursued when the detectors suffer from a not - suitable radiopurity level in the sensitive target - material and in the surroundings .
those activities generally apply a large number of cut procedures to the data ; each one is affected by non - negligible systematic errors which are usually not suitable quantified . as an example , the xenon-10 experiment applies more than 10 different cuts to the data ; the experiment collects @xmath1 events but only 10 are claimed to survive to the cuts and handling procedures @xcite .
very high reduction factors following applied multiple cuts are dangerous because of the difficult precise estimate of all the involved systematics . for example , it has been shown in ref .
@xcite that zeplin - i has claimed a sensitivity 3 orders of magnitude larger than the one properly obtained when accounting for systematics .
thus , the robustness of some results appeared in the `` race for the best exclusion - plot '' ( valid just in a single set of assumptions for a certain kind of wimp ) should be considered `` cum grano salis '' .
for example , for some liquid noble gas set - ups , apart from the robustness of the applied cuts themselves , the very low energy scale and the energy threshold are determined by extrapolating from calibrations at much higher energy and applying some kind of corrections for relevant non - uniformity of the detector s response ; with a light yield of about 2.2 photoelectrons / kev and not specific calibrations even @xmath2 1.5 - 2 kev electron equivalent is claimed as energy threshold @xcite .
considering the energy threshold dependence of the exclusion plots , up to several orders of magnitude differences can be present between claimed and realistic evaluation of the experimental sensitivity .
another crucial aspect is the proper accounting for the existing experimental and theoretical uncertainties in the calculation and in the comparison of experiments using different target materials and approaches . |
radio galaxies exhibit several extended components on the kpc scale , namely : jets , hotspots and lobes .
the lobes are double structures extended over large scales , roughly symmetrical and ellipsoidal lying on both sides of the active nuclei of radio galaxies .
it was first noticed by fanaroff & riley ( 1974 ) that the relative positions of regions of high and low surface brightness in the extended components of extragalactic radio sources are correlated with their radio luminosity .
it was found that nearly all sources with luminosity @xmath5 @xmath6 2 @xmath7 10@xmath8 @xmath9 w hz@xmath10 str@xmath10 were of class i ( i.e. , fri ) while the brighter sources were nearly all of class ii ( i.e. , frii ) .
fris have a surface brightness which is larger toward their cores while friis have it larger toward their edges .
the morphology of radio galaxies turns out to reflect the method of energy transport in the radio source .
fri radio galaxies typically have bright jets in the center , while friis have faint jets but bright hotspots at the ends of their lobes ( e.g. , baum et al .
1995 ) .
the radio emission arising from the lobes is widely interpret as due to synchrotron radiation , while the high - energy radiation has been successfully described as inverse compton emission off the cosmic microwave background ( ic / cmb ) .
moreover , because of the low electron densities in the large lobe volumes the synchrotron self compton ( ssc ) mechanism produces a negligible contribution both at x - ray and at @xmath0-rays ( e.g. , hardcastle et al .
2002 , croston et al .
2005 , hereafter c05 ) .
the radio lobes extension is typically @xmath3 100 kpc implying a timescale for their growth spanning from tens to hundreds of millions of years ( blundell & rawlings 2000 , mullin et al .
if relativistic particles in the lobes are not continuously accelerated , considering the synchrotron cooling time @xmath11 @xmath3 16.4 @xmath12 @xmath13 @xmath14(g ) yr , relativistic electrons with lorentz factor @xmath0 @xmath15 10@xmath2 in magnetic fields of @xmath3 10@xmath16 - 10@xmath17 g do not emit on epochs longer than @xmath3 1 myr , which is shorter than the age of radio lobes ( van der laan & perola 1969 ) .
a population of energetic electrons is required to interpret the radio / millimeter observations of radio galaxy lobes ( e.g. , hardcastle & looney 2008 ) , and recently , the _ wilkinson microwave anisotropy probe _ ( wmap ) and _ fermi _ observations of the centaurus a lobes ( hardcastle et al .
2009 : abdo et al .
in addition , the _ chandra _ , _ xmm - newton _ and _ suzaku _ observations of radio galaxy lobes ( isobe et al . 2002 , 2005 , 2009 ,
kataoka et al .
2003 , c05 ) show that their ic / cmb spectral energy distribution ( sed ) is rising in the x - rays and consequently a significant fraction of their high - energy emission is expected to be radiated in the hard x - rays and at low energy @xmath0-rays .
following the idea of bergamini et al .
( 1967 ) , the ic / cmb process of extended structures in radio galaxies could provide a significant contribution to the diffuse extragalactic @xmath0 ray background ( egb ) in the kev to mev energy range .
the contribution to the egb of kpc - scale jets in the case of fris , from ic scattering of starlight photons by the ultrarelativistic electrons , has been estimated by stawarz et al .
these estimates show that this emission contributes @xmath3 1% of the egb at gev energies , and it is less relevant at mev energies , where the contribution of the ic / cmb emission peaks . in this letter
, we adopt a statistical acceleration scenario for the lobes in friis , due to turbulent magnetic fields , that balances the radiative cooling and keeps electron energies up to @xmath0 @xmath3 10@xmath4 - 10@xmath2 ( e.g. , pacholczyk & scott 1976 , eilek & shore 1989 ) , in agreement with the recent _ fermi _ observations . a similar mechanism has been proposed by hardcastle et al .
( 2009 ) and osullivan et al .
( 2009 ) to identify radio galaxy lobes as possible accelerators for the ultra high - energy cosmic rays ( e.g. , hillas 1984 ) .
finally , we consider a template sed of the emission from the friis lobes to estimate their contribution to the egb .
we use cgs units and a flat cosmology with @xmath18 km s@xmath10 mpc@xmath10 , @xmath19 and @xmath20 ( dunkley et al . 2009 ) .
the relativistic electrons in radio galaxy lobes are subject to two main radiative loss processes : the synchrotron radiation and the ic / cmb .
the first mechanism is responsible for the lobe radio emission , while the latter describes their observed high - energy emission in the x - rays , and recently , also in the @xmath0-rays as in the nearby case of centaurus a ( abdo et al .
2010a ) . to clarify these radiative processes
it is interesting to consider the case of a single relativistic electron of lorentz factor @xmath0 .
the total power emitted via synchrotron and ic / cmb radiation is : @xmath21 where @xmath22 is the lobe magnetic field while @xmath23 = 3.26 @xmath12 10@xmath17@xmath24 g , the value of the magnetic field that has the same energy density as the cmb at redshift @xmath25 ( e.g. , murgia et al .
the @xmath26 ratio identifies which emission process , synchrotron or ic / cmb , is more relevant for the electron radiative cooling .
ratio ( upper panel ) for the friis lobes in the sample of c05 .
the dashed red line indicates the case @xmath26 = 1 , when synchrotron and ic / cmb losses have the same relevance .
b ) the equivalent spherical radius of the lobe volume , @xmath27 , as a function of redshift estimated for the same sample.,width=332,height=283 ] in figure 1(a ) , we report the @xmath26 ratio and @xmath27 , the equivalent spherical radius of the lobe volume , with respect to @xmath25 , for the sample of frii lobes detected in the x - rays ( c05 ) , where @xmath22 has been estimated by fitting the radio - to - x - ray lobe seds .
we note that the ratio @xmath26 is of order a few , in particular for the sources with larger @xmath27 at low redshift ( i.e. @xmath280.5 ) , for which the cmb energy density exceeds the estimate of the magnetic field energy density ( figure 1(b ) ) .
seventy - five percent of the considered frii lobes have @xmath26 in the range 0.72 ( figure 1a ) , suggesting that in these sources the radiative losses for synchrotron and for ic / cmb have similar importance .
this strongly suggests that the luminosity ratio of the two sed components @xmath29 is @xmath3 1
. however , these estimates of the @xmath26 ratio for radio galaxy lobes are dependent on the choice of the particle energy distribution ( ped ) adopted .
the restricted energy ranges , where lobe emission is detected , do not allow us to estimate the ped shape and to constrain the @xmath29 accurately , introducing uncertainties in the estimates of the @xmath26 .
another relevant process for particle energy losses in lobes is adiabatic expansion .
these losses could play a significant role during the evolution of radio lobes ( e.g. , longair et al .
assuming a self similar expansion scenario ( e.g. , matthews & scheuer 1990 , kaiser & alexander 1997 ) , the adiabatic losses are relevant in the initial phase of the lobe evolution , as in the case of young , compact radio galaxies . during the final phase of the lobe expansion , when the lobe pressure is in equilibrium with that of the external medium , losses are dominated by radiative processes , e.g. , via ic / cmb or synchrotron emission ( if @xmath26 is still @xmath30 1 ) .
friis , with lobe size @xmath15 30 - 50 kpc , can be considered at the end of their evolution , when the adiabatic expansion losses are negligible ( e.g. , pacholczyk & scott 1976 ) .
acceleration by turbulent magnetic fields ( due to electron scattering with magnetized plasma irregularities ) via second order _ fermi _
mechanisms , could be considered as a possible mechanism to compensate both the synchrotron and the ic radiative losses ( e.g. , eilek & shore 1989 , osullivan et al . 2009 , hardcastle 2010 ) . adopting this scenario , the acceleration timescale via statistical mechanism
can be written as @xmath31 @xmath32 @xmath33 , where @xmath34 is the characteristic length scale of the magnetic field inhomogeneities ( i.e. , the mean free path of the electrons between collisions ) and @xmath35 = @xmath36 is the ratio of the alfvn velocity @xmath37 and the speed of light @xmath38 . consequently , according to kardashev ( 1962 ) , the energy gain by systematic statistical acceleration , for a single relativistic electron in a turbulent magnetic field @xmath22 is : @xmath39 where @xmath40 is the energy density of the turbulent magnetic field , @xmath41 is the plasma density and @xmath42 and @xmath43 the electron and the proton mass , respectively ( e.g. , tsytovich 1966 , eilek & shore 1989 ) .
we can estimate the maximum lorentz factor of the electrons in the lobes , balancing the acceleration energy gain ( equation ( 2 ) ) with the radiative losses ( equation ( 1 ) ) : @xmath44 that is only dependent on the length scale , @xmath34 , on the plasma density , @xmath45 , and on the @xmath26 ratio .
lobes are usually assumed to be close to the equipartition condition between magnetic field and electrons , implying @xmath46 @xmath47 @xmath3 10@xmath16 g ( e.g. , hardcastle et al .
2002 ) . however , the recent case of pictor a ( migliori et al . 2007 ) or similar friis ( e.g. , brunetti et al . 2002 ,
isobe et al .
2005 , c05 ) suggest that this requirement is not very tight indicating that for a lobe with @xmath25 between 0.5 and 1 , as the majority of frii lobes are detected in x - rays , @xmath26 is of the order of unity ( see figure 1a ) .
the mean free path , until an electron collides with a magnetic inhomogeneity in a turbulent plasma , must be smaller than the source size and it has been estimated to be of the order of @xmath3 1 kpc ( e.g. , lacombe 1977 ) .
consequently , substituting @xmath34 @xmath3 10@xmath48@xmath27 @xmath32 10@xmath49 cm into equation ( 3 ) , and assuming @xmath26 @xmath3 1 , @xmath50 is of the order of 10@xmath4 - 10@xmath2 , as required to interpret the @xmath0-ray emission of centaurus a lobes as ic / cmb and in agreement with the estimate of osullivan et al .
this is consistent the radio / millimeter observations of radio galaxies ( e.g. , hardcastle & looney 2008 ) , that imply the presence of a population of high - energy electrons in their lobes ( i.e. , @xmath51 10@xmath4 - 10@xmath2 ) .
our scenario is also in agreement with that proposed by pacholczyk & scott ( 1976 ) for the in situ re - acceleration of extended structures in radio galaxies .
we note that electrons with energies @xmath0 @xmath3 10@xmath52 - 10@xmath53 have synchrotron cooling times @xmath11 @xmath3 10@xmath54 - 10@xmath55 yr and are not expected to be present in radio lobes at the epochs of about few myrs ( which is the typical lobe age ) .
this is in agreement with the non - detection of lobes at ir and optical frequencies ( i.e. , @xmath56 @xmath3 10@xmath57hz ) , where the synchrotron emission in a magnetic field of about @xmath58 - @xmath59 g of electrons with @xmath0 @xmath3 10@xmath52 - 10@xmath53 should be seen . [
sec : egb].,width=328,height=283 ]
the cmb is a blackbody exhibiting the peak frequency of the sed at @xmath60 @xmath3 1.6 @xmath12 10@xmath61 hz ( adopting the wien law ) .
consequently , the expected peak frequency of the ic emission in thomson regime is @xmath62 .
low energy electrons at @xmath63 radiate via ic / cmb in the x - ray band , while electrons with @xmath1 - 10@xmath2 emit in the mev - gev energy range .
the emission arising from highest energy electrons produces the tail of the ic / cmb sed component and as such its intensity is not expected to be extremely large making the source undetectable by _ fermi _ , unless the radio galaxy lies at very low redshift .
as previously anticipated this high - energy radiation could contribute to the diffuse egb . to take this contribution into account
, we built a template sed of the lobe emission from the radio to @xmath0-rays .
the template has been evaluated assuming a smooth log - normal ped : @xmath64 , ( massaro et al .
this distribution has been suggested as intrinsic shape of lobe and hotspots ( e.g. , katz - stone & rudnick 1994 ) . the parameters considered to calculate our sed template are reported in table 1 and it is shown in figure 2a .
llcc parameter & symbol & units & template + redshift & z & & 0.2 + ped curvature & r & & 0.7 + ped energy peak & @xmath65 & & 7@xmath1210@xmath54 + ped min energy & @xmath66 & & 10@xmath67 + ped max energy & @xmath50 & & @xmath68 + ped density & n & @xmath69 & 10@xmath70 + luminosity ratio & @xmath29 & & 1 + equivalent lobe radius & @xmath27 & kpc & 45
+ + we did not assume that the ic / cmb emission is beamed , since there is no evidence of relativistic bulk motion in lobes .
we chose the value of @xmath27 to be consistent with the typical value of the c05 sample ( see figure 1(b ) ) , and @xmath65 to have the peak of the synchrotron emission in the radio band at @xmath71 ghz .
this is justified by the case of 3c 219 at @xmath25 = 0.1744 ( comastri et al . 2003 ) , for which high frequency radio observations are available .
the predicted ir flux of the template at 3.6@xmath72 m is @xmath32 7 @xmath7 10@xmath73 ergs@xmath10cm@xmath48 , in agreement with the non - detected emission at infrared and optical frequencies .
the template is in agreement with the observed radio - to - x - ray luminosities measured in the lobe sample of c05 ( see figure 2(b ) and with the models adopted for the x - ray detected lobes of other friis ( e.g. , 3c 219 , comastri et al .
2003 , 3c 207 , brunetti et al .
2002 , 3c 452 , 3c 98 , 3c 326 isobe et al .
2002 , 2005 , 2009 , pictor a , migliori et al . 2007 ) .
we calculate the sed template with the luminosity ratio between the synchrotron and the ic / cmb component @xmath29 of the order of unity ( figure 2a ) .
this corresponds to the assumption that the energy losses for the radiative processes considered are equivalent , similar condition of @xmath26 @xmath3 1 , in agreement with the observations of friis ( see sect .
2 ) . in addition , the x - ray spectral index is consistent with the typical slope observed for the few lobes for which a detailed spectral analysis exists ( hardcastle et al .
2002 , c05 , isobe et al .
2002 , 2005 , 2009 ) .
the estimate of the total diffuse flux arising from the frii radio galaxy population can be evaluated adopting the following relation , as reported in ajello et al .
( 2008 , 2009 ) : @xmath75 where @xmath76 is the diffuse flux in units of photons cm@xmath48 s@xmath10 sr@xmath10 mev@xmath10 , @xmath77 is the luminosity function ( lf ) , @xmath78 is derived from our sed template of the lobes in the observer s frame at energy @xmath79 and @xmath80 is the comoving volume element per unit redshift and unit solid angle ( hogg 1999 ) .
the knowledge of the source lf is crucial to estimate correctly this contribution .
indeed , it gives the space density of objects with luminosity @xmath81 at redshift @xmath25 .
we adopt the frii radio galaxy lf at 15ghz ( cara & lister 2008 ) .
the sed template used in equation ( 4 ) is renormalized such that at a given redshift @xmath25 the luminosity in the native radio band of the lf is @xmath81 .
moreover , the ic component is multiplied by an additional factor @xmath82 to account for the fact that the energy density of the cmb scales like @xmath83 .
the luminosity of the lf of cara & lister ( 2008 ) refers to the luminosity of the frii core while our sed template applies to the lobe emission . to convert the core luminosity into a proxy for the lobe luminosity at 15ghz we have used the available observations of frii ( c05 )
we have found that rescaling the core luminosity by a factor @xmath84 ( and re - normalizing the sed template as described above ) produces seds that reproduce well the available data on friis for the range of source luminosities and redshifts reported in c05 .
we randomly extracted sources from our model and selected only those with flux at 1kev @xmath15 10@xmath85erg cm@xmath48 s@xmath10 and @xmath86 .
as shown in figure ( 2 ) our lf model is able to reproduce successfully the range of radio and x - ray luminosities of the c05 sample ( see figure 2(b ) ) .
moreover , we performed an additional test integrating the lf coupled to our sed model over luminosity and redshift and we counted how many objects would be detectable in an all - sky survey above a flux of 10@xmath85erg cm@xmath48 s@xmath10 in the x - ray band .
the number of detectable sources is @xmath3100 and they display a typical ( average ) x - ray luminosity of @xmath87erg cm@xmath48 s@xmath10 .
these numbers appear to be consistent with the _ chandra _ and _ xmm - newton _ observations ( e.g. , c05 ) . as an example , adopting a value for @xmath88 of 10@xmath10 or 10@xmath89 would produce @xmath34000 or @xmath31 observable sources in the entire sky .
both these scenarios seem very unlikely given the observations described above .
we estimated the uncertainties on @xmath88 adopting the following approach . to set a lower limit on k
, we required that the number of detectable lobes all - sky ( estimated through equation ( 4 ) ) at 1kev , with flux @xmath15 10@xmath85erg cm@xmath48 s@xmath10 matches ( within 1@xmath90 ) the number of sources in the c05 sample .
this condition is satisfied for @xmath91 .
then , to set the upper limit , we request that the density of sources observable at x - rays ( same criteria as above ) matches the frii density at 178mhz in the complete sample of mullin et al .
this condition is satisfied for @xmath92 .
3 shows the contribution to the diffuse egb emission arising from the extended structures of frii lobes and its uncertainty .
it is apparent that lobes give a significant , @xmath310% contribution to the egb in the mev energy range .
in this letter , we adopted a statistical acceleration scenario based on turbulent magnetic fields to show that balancing the particle energy gain with the radiative losses , the expected lorentz factor is of the order of @xmath0 @xmath3 10@xmath4 - 10@xmath2 for electrons in the frii lobes , in agreement with the radio / millimeter observations of lobes in radio galaxies ( e.g. , hardcastle & looney 2008 ) .
this has been estimated assuming that the acceleration length scale is comparable to the mean free path in the lobe not far from the equipartition condition , supported by the recent x - ray analyses of c05 .
the above scenario is able to justify the presence of high - energy electrons responsible for the @xmath0-ray emission arising from the radio galaxy lobes , via ic / cmb , as the _ fermi
_ observations of the nearby centaurus a ( abdo et al .
2010a ) .
considering the ic / cmb scattering as the main radiative process for the emission in the mev energy range , we estimated the contribution of the friis to the diffuse egb .
we found that the peak of this diffuse emission lies at @xmath3 1 mev and the radiation arising from lobes could contribute @xmath3 10% of the egb close to this energy .
so far , the lobes in radio galaxies were only observable at radio frequencies and at high energies in the x - rays .
this implies that the shape of the emitting ped is uncertain .
consequently , the estimates of the lobe magnetic field and the possibility of identifying which is the most relevant radiative process , synchrotron or ic emission are not well constrained .
however , current models adopted to describe the sed of lobes in friis assume the presence of high - energy electrons up to @xmath51 10@xmath4 - 10@xmath2 ( e.g. , c05 ) . to test this hypothesis , very deep infrared and/or hard x - ray observations , with sufficient spatial resolution are necessary .
if _ fermi _ will detect @xmath0-ray emission arising from nearby frii radio galaxies , these observations will be crucial to confirm the presence of high - energy electrons in radio lobes and to constrain their sed .
we thank the anonymous referee for the constructive suggestions that have improved and strengthened the manuscript .
we are grateful to m. murgia , d. harris and a. cavaliere for fruitful suggestions and to j. finke , l. stawarz and t. cheung for their comments . f. massaro acknowledges the foundation blanceflor boncompagni - ludovisi , nee bildt for the grant awarded him in 2010 .
the work at sao is supported by the nasa grant nnx10ad50 g .
m. ajello acknowledges support by nasa grant nnh09zda001n .
adbo a. a. et al .
2010a sci , 328 , 725 adbo a. a. et al .
2010b apj , 720 , 435 ajello , m. et al .
2008 apj , 689 , 666 ajello , m. et al .
2009 apj , 699 , 603 baum , stefi a. , zirbel , esther l. , odea , christopher p. 1995 , apj , 451 , 88 bergamini , r. , londrillo , p. , setti , g. 1967 ncimb , 52 , 495 blundell , k. m. & rawlings , s. 2000 aj , 119 , 1111 brunetti , g. , bondi , m. , comastri , a. , setti , g. 2002 a&a , 381 , 795 cara , m. & lister , m. l. 2008 apj , 674 , 111 comastri , a. , brunetti , g. , dallacasa , d. , bondi , m. , pedani , m. , setti , g. 2003 mnras , 340l , 52 croston , j. h. , hardcastle , m. j. , harris , d. e. , belsole , e. , birkinshaw , m. , worrall , d. m. 2005 apj , 626 , 733 dunkley , j. et al .
2009 apj , 701 , 1804 eilek j. a. & shore , s. n.1989 apj , 342 , 187 fanaroff , b.l . & riley j.m .
1974 , mnras , 167 , 31 fukada , y. et al . 1975 ,
nature , 254 , 398 hillas , a. m. 1984 ara&a , 22 , 425 hardcastle , m. j. , birkinshaw , m. , cameron , r. a. , harris , d. e. , looney , l. w. , worrall , d. m. 2002 apj , 581 , 948 hardcastle , m. j. & looney l. w. 2008 mnras , 388 , 176 hardcastle , m. j. , cheung , c. c. , feain , i. j. , stawarz , l. 2009 mnras , 393 , 1041 hardcastle , m. j. 2010 mnras , 405 , 2810 hogg , d. w. 1999 , arxiv : astroph/9905116v4 isobe , n. , tashiro , m. , makishima , k. , iyomoto , n. , suzuki , m. , murakami , m. m. , mori , m. , abe , k. 2002 apj , 580 , l111 isobe , n. , makishima , k. , tashiro , m. , hong , s. 2005 apj , 632 , 781 isobe , n. , tashiro , m. s. , gandhi , p. , hayato , a. , nagai , h. , hada , k. , seta , h. , matsuta , k. 2009 apj , 706 , 454 kaiser , c. r. & alexander , p. 1997
mnras , 286 , 215 kardashev , n. s. 1962 sva , 6 , 317 kataoka , j. , leahy , j. p. , edwards , p. g. , kino , m. , takahara , f. , serino , y. , kawai , n. , martel , a. r. 2003 a&a , 410 , 833 katz - stone , d. m. & rudnick , l. 1994 apj , 426 , 116 lacombe , c. 1977 a&a , 54 , 1 longair , m. s. , ryle , m. , scheuer , p. a. g. 1973 mnras , 164 , 243 massaro , e. , perri , m. , giommi , p. , nesci , r. , verrecchia , f. 2004 , a&a , 422 , 103 matthews , a. p. & scheuer , p. a. g. 1990 mnras , 242 , 616 migliori , g. , grandi , p. , palumbo , g. c. , brunetti , g. , stanghellini , c. 2007 apj , 668 , 203 murgia , m. , fanti , c. , fanti , r. , gregorini , l. , klein , u. , mack , k .- h . , vigotti , m. 1999 a&a , 345 , 769 mullin , l. m. , riley , j. m. , hardcastle , m. j. 2008 mnras , 390 , 595 osullivan , s. , reville , b.;,taylor , a. m. 2009 mnras , 400 , 248 pacholczyk , a. g. & scott , j. s. 1976 apj , 203 , 313 stawarz , l. , kneiske , t. m. & kataoka , j. 2006 apj , 637 , 693 tsytovich , v. n. 1966 , sov . phys .
, 9 , 370 van der laan , h. & perola , g. c. 1969 a&a , 3 , 468 watanabe , k. 1997 , in aip conf .
410 , proc .
fourth compton symposium , ed . c. d. dermer ,
m. s. strickman , & j. d. kurfess ( melville : aip ) , 1223 weidenspointner , g. , et al .
2000 , in aip conf .
510 , american institute of physics conference series , ed .
m. l. mcconnell & j. m. ryan ( melville : aip ) , 467 | the recent discovery of the @xmath0-ray emission from the lobes of the closest radio galaxy centaurus a by _ fermi _ implies the presence of high - energy electrons at least up to @xmath1 - 10@xmath2 .
these high - energy electrons are required to interpret the observed @xmath0-ray radiation in terms of inverse compton emission off the cosmic microwave background ( ic / cmb ) ; the widely accepted scenario to describe the x - ray emission of radio galaxy lobes . in this letter , we consider the giant radio lobes of fr ii radio galaxies showing that it is possible to maintain electrons at energies @xmath0 @xmath3 10@xmath4 - 10@xmath2 , assuming an acceleration scenario ( driven by turbulent magnetic fields ) that compensates the radiative losses .
in addition , we consider the contribution to the diffuse extragalactic @xmath0-ray background due to the ic / cmb emission of friis lobes showing its relevance in the kev to mev energy range . |
self - consistent mean - field calculations using effective interactions have been very successful in describing the properties and the dynamics of a wide range of nuclei @xcite .
one of the most popular effective interactions are zero - range interactions of the skyrme type because of the fast numerical computations which can thus be performed @xcite .
even though negele and vautherin @xcite showed a long time ago how to obtain effective interactions using nuclear many - body methods , a more empirical approach has been usually followed .
a specific parametric form of the effective interaction is postulated and the unknown parameters are determined _ a posteriori _ so as to reproduce a set of nuclear data selected according to a specific purpose .
the non - uniqueness of the fitting procedure has thus lead to a large number of different parametrizations .
some of them may yield very different predictions when applied outside the domain where they were fitted @xcite .
this situation is particularly unsatisfactory for nuclear astrophysical applications which require the knowledge of nuclear masses for nuclei so neutron rich that there is no hope of measuring them in the foreseeable future ; such nuclei play a vital role in the r - process of nucleosynthesis @xcite and are also found in the outer crust of neutron stars @xcite .
extrapolations far beyond the neutron drip line are required for the description of the inner crust of neutron stars where nuclear clusters are embedded in a sea of superfluid neutrons @xcite .
such environments can not be reproduced in the laboratory .
however many astrophysical phenomena observed in neutron stars are precisely related to the physics of these outer layers @xcite .
the need for more reliable extrapolations of these nuclear models has motivated recent efforts to construct non - empirical effective interactions and more generally microscopic nuclear energy density functionals @xcite .
unfortunately such _ ab initio _ nuclear energy density functionals able to reproduce existing experimental nuclear data with the same degree of accuracy as phenomenological interactions are not yet available . for the time being , a reasonable approach is to fit these phenomenological interactions with as many available nuclear data as possible , not only from experiments but also from realistic calculations in infinite homogeneous nuclear matter .
very accurate nuclear mass models based on the hartree - fock - bogoliubov ( hfb ) method with zero- and finite - range effective forces have thus been recently developed @xcite .
in particular in our model hfb-17 @xcite , we have achieved our best fit ever to essentially all the available experimental mass data , the rms deviation for the set of 2149 measured masses of nuclei with @xmath0 and @xmath1 8 @xcite being only 0.581 mev .
significant improvements compared to our previous models have been made possible by a better treatment of pairing correlations first introduced in ref .
@xcite and extended in ref .
@xcite . the effective interaction used in the pairing channel
was constrained to reproduce the @xmath2 pairing gaps as obtained by microscopic calculations using realistic nucleon - nucleon interactions .
we have recently applied these effective forces to study the properties of the neutron superfluid phase in the inner crust of neutron stars @xcite .
the effective pairing interaction associated with our nuclear mass models requires the evaluation of one dimensional integrals .
although its numerical implementation is straightforward , using such interaction could become computationally costly for large scale calculations such as the determination of fission barriers . in this paper , a more tractable expression of the pairing interaction is presented by calculating the underlying integrals analytically .
we will consider effective interactions of the skyrme type in the particle - hole channel .
but results can be easily generalized to other kinds of interactions .
the pairing interaction that we consider here acts only between nucleons of the same charge state @xmath3 ( @xmath4 or @xmath5 for neutron or proton respectively ) and is given by @xmath6~\delta(\pmb{r_{ij}})\ , , \ ] ] where @xmath7 is the two - body spin - exchange operator , @xmath8 and @xmath9 . due to the zero range of this interaction , a cutoff has to be introduced in the gap equations in order to avoid divergences ( for a review of the various prescriptions , see for instance ref .
@xcite ) . in this work
we include all single - particle states whose energy lies below @xmath10 , where @xmath11 is the chemical potential and @xmath12 is an energy cutoff . although microscopic calculations in semi - infinite nuclear matter suggest that pairing in nuclei is a surface phenomenon @xcite , the density dependence of the pairing force is still poorly known .
it is generally assumed that @xmath13 $ ] depends only on the isoscalar density @xmath14 and is often parametrized as @xcite @xmath15 = v_{\pi q}^\lambda\left(1-\eta_q \left(\frac{\rho}{\rho_0}\right)^{\alpha_q}\right ) \
, , \ ] ] where @xmath16 is the nuclear saturation density while @xmath17 , @xmath18 and @xmath19 are adjustable parameters .
the superscript @xmath20 on @xmath17 is to remind that the pairing strength depends very strongly on the cutoff used in the gap equations ( in principle changing the cutoff modifies also the other parameters but the effects are usually found to be small ) .
effective interactions with @xmath21 ( @xmath22 ) have been traditionally refered as volume ( surface ) pairing .
the parameters in eq .
( [ eq.2 ] ) are usually fitted directly to experimental data .
the standard prescription is to adjust the value of the pairing strength @xmath17 to the average gap in @xmath23sn @xcite .
however this does not allow an unambiguous determination of the remaining parameters @xmath18 and @xmath19 .
systematic studies of nuclei seem to favor a so - called mixed pairing with @xmath24 and @xmath25 @xcite .
even though such forces have been widely applied in nuclear structure calculations with some success , they lack a direct connection with realistic nucleon - nucleon forces .
the reliability of these forces far beyond the domain where they were fitted is therefore not guaranteed .
in particular pairing forces fitted only to finite nuclei generally yield unrealistic pairing gaps in infinite nuclear matter @xcite , thus rendering their application to neutron - star crusts unreliable .
garrido et al .
@xcite proposed to determine the parameters of the pairing interaction in eq .
( [ eq.2 ] ) by fitting the @xmath2 pairing gaps in infinite symmetric nuclear matter as obtained by the realistic paris potential in the bcs approximation .
the parametric form eq .
( [ eq.2 ] ) has been recently extended by margueron et al .
@xcite who introduced an isospin dependence in the pairing strength in order to reproduce the @xmath2 pairing gaps in both symmetric nuclear matter and pure neutron matter as obtained from brueckner calculations @xcite .
this approach assumes that the pairing _ interaction _ between two nucleons inside a nucleus is locally the same as the pairing interaction between two nucleons in infinite uniform matter .
this should not be confused with the so - called `` local density approximation '' which supposes that the pairing _ field _
@xmath26 appearing in the hfb equations ( see for instance appendix a of ref .
@xcite ) is locally the same in finite nuclei and in infinite nuclear matter .
even though the coupling to surface vibrations is expected to contribute to pairing @xcite , a local pairing theory seems a reasonable first step @xcite .
finite size effects can be subsequently taken into account by introducing density gradient terms in the expression of the effective pairing force as suggested in ref .
@xcite in the general framework of the density functional theory .
effective pairing forces with density gradients have been implemented in ref .
@xcite .
the drawback of using a phenomenological pairing force in order to fit microscopic pairing gaps in infinite uniform matter is that the functional form of the associated pairing strength is not _ a priori _ known . in particular different pairing gaps require different expressions of @xmath13 $ ] which can hardly be guessed due to the highly non - linear character of pairing correlations @xcite .
it should also be remarked that the parameters of the pairing force depend on the effective interaction used in the particle - hole channel since effective masses appear in the gap equations .
fitting the parameters of the effective interaction to essentially all experimental nuclear masses both in the particle - hole and the particle - particle channels while simultaneously reproducing microscopic pairing gaps in infinite uniform matter would thus be an extremely onerous numerical task .
actually as shown in refs .
@xcite , the density- and isospin dependence of the pairing strength can be calculated exactly for any given @xmath2 pairing - gap function @xmath27 $ ] by solving directly the hfb equations in infinite uniform matter yielding @xmath28=-\frac{8\pi^2}{i_q(\rho_n,\rho_p)}\left(\frac{\hbar^2}{2 m_q^*(\rho_n,\rho_p)}\right)^{3/2}\ ] ] where @xmath29 is the nucleon effective mass and @xmath30 the expression of @xmath29 for skyrme forces can be found for instance in appendix a of ref .
@xcite . in eq .
( [ eq.4 ] ) , @xmath31 is a reduced chemical potential ( @xmath32 being the mean field potential ) which can be obtained by imposing the conservation of the nucleon particle number density @xmath33 however it is usually a very good approximation to replace @xmath34 in eq .
( [ eq.4 ] ) by the fermi energy @xmath35 where @xmath36 is the fermi wave number .
the approximation @xmath37 holds provided @xmath38 which is typically the case @xcite . using eq .
( [ eq.3 ] ) instead of a phenomenological expression such as eq ( [ eq.2 ] ) guarantees that in infinite uniform matter the microscopic pairing gaps @xmath39 will be exactly reproduced .
it also ensures that the pairing strength is suitably renormalized for any change in the energy cutoff .
the price to be paid is the evaluation of the one dimensional integral , eq ( [ eq.4 ] ) , for the neutron density @xmath40 and proton density @xmath41 at all points inside the nucleus . the computational cost becomes significant in fully self - consistent 2d and 3d calculations @xcite .
however we will now show that these numerical integrations can be avoided by making use of the weak - coupling approximation .
the integrand in eq .
( [ eq.4 ] ) is all the more peaked around @xmath42 as @xmath43 is small compared to @xmath44 .
this suggests to expand the integrand in powers of @xmath45 leading to @xmath46 where the dimensionless coefficients @xmath47 are defined by @xmath48 in the weak - coupling approximation @xmath49 and @xmath50 , only the first coefficient in eq .
( [ eq.7 ] ) is usually retained , i.e. @xmath51 .
this approximation is equivalent in taking a constant density of single - particle states in the gap equations . calculting the integral @xmath52 and keeping lowest order terms in @xmath53 and @xmath54 , eq .
( [ eq.8 ] ) thus yields @xmath55 inserting eq .
( [ eq.9 ] ) in eq .
( [ eq.3 ] ) and solving for @xmath43 leads to the familiar expression ( note that @xmath56 ) @xmath57 even though the weak - coupling approximation provides good results in the case of conventional bcs superconductivity @xcite , it is less accurate in the nuclear context because of the large number of states involved in the pairing mechanism
. nevertheless the higher - order coefficients in eq .
( [ eq.8 ] ) can be easily evaluated . calculating the integrals @xmath47 , keeping as before lowest - order terms in @xmath53 and @xmath54 , and summing all coefficients
, we find @xmath58 - 4\ , , \ ] ] where @xmath59 .
adding eqs .
( [ eq.9 ] ) and ( [ eq.10 ] ) in eq .
( [ eq.7 ] ) leads to @xmath60\ ] ] with @xmath61 the highly - non linear character of the pairing phenomenon is evident in eqs .
( [ eq.3 ] ) and ( [ eq.11 ] ) .
these equations also show that the density dependence of the pairing strength is intimately related to the choice of the pairing cutoff . the contribution of the latter is entirely contained in the function @xmath62 .
it can also be seen from eqs .
( [ eq.3 ] ) and ( [ eq.11 ] ) that the pairing strength vanishes whenever the pairing gap @xmath43 goes to zero at _ finite _ density , as expected .
note that as a by - product we have also obtained a more accurate expression of the pairing gap .
substituting eq .
( [ eq.11 ] ) in eq .
( [ eq.3 ] ) yields @xmath63 where @xmath59 .
$ ] , as defined by eq.([eq.3 ] ) , vs nucleon number density @xmath64 in symmetric nuclear matter using the effective interaction underlying the nuclear mass model hfb-17 @xcite .
the symbols are the results of numerically integrating eq .
( [ eq.4 ] ) . the dashed and solid lines were obtained using either eq .
( [ eq.9 ] ) or eq .
( [ eq.11 ] ) respectively . ]
but for pure neutron matter .
] we have tested the validity of the analytical expression given by eq .
( [ eq.11 ] ) as compared to the numerical integration of eq .
( [ eq.4 ] ) using the effective interactions underlying our hfb-17 nuclear mass model @xcite .
the pairing force used in this model was adjusted so as to reproduce the @xmath65s@xmath66 pairing gaps in both symmetric nuclear matter and pure neutron matter as obtained from brueckner calculations using the argonne @xmath67 nucleon - nucleon potential @xcite .
the pairing cutoff was fixed at @xmath68 mev . as can be seen in figures [ fig1 ] and [ fig2 ] , eqs .
( [ eq.3 ] ) and ( [ eq.11 ] ) provide a very good estimate of the pairing strength for all densities and different isospin asymmetries .
the figures also show that keeping only the first term in eq .
( [ eq.7 ] ) with @xmath52 given by eq .
( [ eq.9 ] ) , leads to fairly good results except at low densities . at very low densities @xmath69 , both the pairing gap @xmath43 and the chemical potential
vanish so that the integral @xmath70 is then simply given by @xmath71 .
it is easily seen from eq .
( [ eq.3 ] ) that the pairing strength remains finite in this limit and is given by @xmath72=-\frac{4\pi^2}{\sqrt{\varepsilon_\lambda}}\left(\frac{\hbar^2}{2 m_q}\right)^{3/2 } \
, , \ ] ] assuming that @xmath73 .
the weak - coupling expression , given by eqs . ( [ eq.3 ] ) and ( [ eq.11 ] ) , tends to the exact limit as shown in figure [ fig3 ] , whereas eq .
( [ eq.9 ] ) leads to a divergence ( the corresponding pairing strength lies far below the range shown in figure [ fig3 ] ) .
the reason for this discrepancy lies in the underlying assumption of a constant density of s.p .
states , which is strongly violated at low energies @xmath74 where the density of states decreases as @xmath75 .
the value of the pairing cutoff @xmath12 can be a priori arbitrarily chosen .
however it has been argued @xcite that in the limit @xmath69 , the pairing strength should coincide with the bare force in the @xmath65s@xmath66 channel , which in turn is determined by the experimental @xmath65s@xmath66 nucleon - nucleon phase shifts . according to eq .
( [ eq.15 ] ) , specifying @xmath76 $ ] is equivalent to specifying the pairing cutoff @xmath12 . as shown in fig . 1 of ref .
@xcite , the optimal value of the cutoff is @xmath77 mev ( note that @xmath12 is half the cutoff used in ref .
@xcite ) . using the experimental phase shifts
would thus remove the only free remaining parameter . on the other hand ,
choosing such a low cutoff can deteriorate the precision of the weak - coupling approximation , since eq .
( [ eq.11 ] ) was obtained assuming @xmath78 .
we have therefore calculated the relative error between the pairing strength calculated numerically and that obtained from eqs .
( [ eq.3 ] ) and ( [ eq.11 ] ) , for different values of the pairing cutoff .
results are shown in figs .
[ fig4 ] and [ fig5 ] for symmetric nuclear and pure neutron matter respectively .
we have checked that the differences come solely from the weak - coupling approximation , and not from the numerical method used to solve eq .
( [ eq.4 ] ) .
it can be seen that for the effective interactions underlying our hfb-17 nuclear mass model @xcite , the weak - coupling approximation is quite accurate since the largest relative error is of a few percent only for @xmath79 mev .
the precision is higher in neutron matter than in symmetric matter because the @xmath65s@xmath66 pairing gap of neutron matter is smaller than that of symmetric matter .
note that the pairing gaps we adopted in our hfb-17 mass model are rather large ( the maximum of the gap function @xmath39 is about @xmath80 mev in symmetric matter and @xmath81 mev in neutron matter ) .
these gaps were obtained from brueckner calculations including medium polarization effects but without self - energy corrections @xcite . when both effects are taken into account , the maximum pairing gaps are much lower @xcite
. the precision of the weak - coupling approximation would thus be even better with such gaps . in any case , if the constraint of reproducing the @xmath65s@xmath66 nucleon - nucleon phase shifts is released , one is free to adjust the cutoff so as to achieve a better precision , as illustrated in figs [ fig4 ] and [ fig5 ] .
$ ] , as defined by eq.([eq.3 ] ) , vs nucleon number density @xmath64 in symmetric nuclear matter ( snm ) and pure neutron matter ( neum ) using the effective interaction underlying the nuclear mass model hfb-17 @xcite .
the symbols are the results of numerically integrating eq .
( [ eq.4 ] ) while the solid line was obtained from eq .
( [ eq.11 ] ) . ] $ ] calculated numerically and that obtained from the weak - coupling approximation , vs nucleon number density @xmath64 .
the different curves correspond to different values of the pairing cutoff @xmath12 indicated in the plot .
calculations were carried out in symmetric nuclear matter using the effective interaction underlying the nuclear mass model hfb-17 @xcite . ] but in neutron matter . ] in the more general context of the nuclear energy density functional theory , the hfb equations can be obtained by minimizing the energy with respect to the normal and pairing density matrices for a fixed average number of neutrons and protons @xcite ( the link between different formulations of the hfb equations is discussed for instance in ref .
all that is needed is therefore the specification of the nuclear energy density functional . the pairing energy density associated with an effective pairing force defined by eq .
( [ eq.1 ] ) , is given by @xmath82 \tilde{\rho}_q(\pmb{r})^2 \ , , \ ] ] where @xmath83 is the so - called local pairing density .
the pairing field appearing in the hfb equations is then defined by @xmath84 because @xmath85 depends on the nucleon densities , it contributes also to the mean - field potentials ( see for instance appendix a of ref .
local pairing functionals can be constructed from infinite nuclear - matter calculations using eqs .
( [ eq.3 ] ) , ( [ eq.11 ] ) and ( [ eq.12 ] ) .
the construction of non - empirical ( and non - local ) pairing functionals using realistic nucleon - nucleon interactions have been discussed in ref .
@xcite ( see also ref.@xcite for a review ) .
in conclusion , we have shown how to construct non - empirical effective contact pairing forces ( or equivalently local nuclear pairing energy density functionals ) using any given @xmath65s@xmath66 pairing - gap functions @xmath39 obtained from microscopic calculations in infinite uniform nuclear matter with realistic nucleon - nucleon potentials .
the resulting parameter - free pairing forces , embodied in eqs .
( [ eq.3 ] ) , ( [ eq.11 ] ) and ( [ eq.12 ] ) , can be easily implemented in nuclear - structure calculations .
these forces could be helpful for understanding the origin of pairing in finite nuclei , and could be used to estimate the importance of bulk contribution as compared to finite - size effects .
alternatively , the analytical solution of the bcs gap equations which we have obtained in eq .
( [ eq.13 ] ) , can be applied to calculate the pairing gaps in nuclear matter for any given contact pairing force ( provided the conditions for the weak - coupling approximation remains valid ) .
even though we have been interested in nuclear pairing , the present results could be easily adapted to other contexts .
this work was financially supported by fnrs ( belgium ) , by the communaut franaise de belgique ( actions de recherche concertes ) and by compstar , a research networking programme of the european science foundation .
the author is grateful to s. goriely and j.m .
pearson for valuable comments .
99 m. bender , p .- h . heenen and p
reinhard , rev .
phys . * 75 * , 121 ( 2003 ) .
j. r. stone and p.g .
reinhard , prog .
. phys . * 58 * , 587 ( 2007 ) .
j. w. negele and d. vautherin , phys .
c * 5 * , 1472 ( 1972 ) .
j. r. stone , j.c .
miller , r .
koncewicz , p.d .
stevenson , m.r .
strayer , phys .
c * 68 * , 034324 ( 2003 ) .
m. arnould , s. goriely , k. takahashi , phys
. rep . * 450 * , 97 ( 2007 ) .
pearson , s. goriely , n. chamel , m. samyn , m. onsi , aip conf .
m. onsi , a. k. dutta , h. chatri , s. goriely , n. chamel , j. m. pearson , phys .
c * 77 * , 065805 ( 2008 ) .
n. chamel and p. haensel,``physics of neutron star crusts '' , living rev .
relativity 11 , ( 2008 ) , 10 .
j. e. drut , r. j. furnstahl , l. platter , prog .
. phys . * 64 * , 120 ( 2010 ) .
n. chamel , s. goriely , and j. m. pearson , phys .
c * 80 * , 065804 ( 2009 ) .
s. goriely , s. hilaire , m. girod , s. peru , phys .
rev . lett.*102 * , 242501 ( 2009 ) .
s. goriely , n. chamel , and j. m. pearson , phys . rev
. lett . * 102 * , 152503 ( 2009 ) .
s. goriely , n. chamel , and j. m. pearson , eur
. j. phys .
a * 42 * , 547 ( 2009 ) .
g. audi , a.h .
wapstra , and c. thibault , nucl .
a729 * , 337 ( 2003 ) .
n. chamel , s. goriely , and j. m. pearson , nucl . phys . *
a812 * , 72 ( 2008 ) .
n. chamel , s. goriely , j.m .
pearson , m. onsi , phys .
c * 81 * , 045804 ( 2010 ) .
t. duguet , k. bennaceur , p. bonche , nucl - th/0508054 , in proceedings of the yitp workshop on new developments in nuclear self - consistent mean - field theories , kyoto , 2005 ( yitp - w-05 - 01 ) , p. b20 .
m. baldo , u. lombardo , e. e. saperstein and m. v. zverev , phys . rep . *
391 * , 261 ( 2004 ) .
g. f. bertsch and h. esbensen , ann . phys . * 209 * , 327 ( 1991 ) .
j. dobaczewski , w. nazarewicz , and t.werner , phys . scr . *
t56 * , 15 ( 1995 ) .
j. dobaczewski , w. nazarewicz , p .-
reinhard , nucl . phys . *
a693 * , 361 ( 2001 ) .
m. samyn , s. goriely , j.m .
pearson , nucl .
a725 ( 2003 ) 69 .
s. takahara , n. onishi , n. tajima , phys .
b * 331 * , 261 ( 1994 ) .
e. garrido , p. sarriguren , e. moya de guerra , p. schuck , phys .
rev.c 60 ( 1999 ) , 064312 .
j. margueron , h. sagawa and k. hagino , phys .
c * 77 * , 054309 ( 2008 ) .
cao , u.lombardo , and p.schuck , phys .
c * 74 * , 064301 ( 2006 ) .
d.m . brink and r.a .
broglia , nuclear superfluidity , cambridge university press ( 2005 ) .
w. satula , phys .
* t125 * , 82 ( 2006 ) .
a. bulgac , phys.rev.c * 65 * , 051305(r ) ( 2002 )
. a. bulgac , y. yu , int.j.mod.phys .
* e13 * , 147 ( 2004 ) .
fayans , s.v .
tolokonnikov , e.l .
trykov , d. zawischa , nucl .
a676 * ( 2000 ) 49 .
t. duguet , phys . rev .
c * 69 * , 054317 ( 2004 ) .
zhang , l.g .
cao , u. lombardo , e.g. zhao , s.g .
zhou , phys .
c * 81 * , 044313 ( 2010 ) .
u. lombardo , superfluidity in nuclear matter , in _ nuclear methods and nuclear equation of state _ , ed . by m. baldo ( world scientific , singapore , 1999 ) , 458 - 510 .
j. bardeen , l. n. cooper , and j. r. schrieffer , phys . rev . * 108 * , 1175 ( 1957 ) .
h. esbensen , g.f .
bertsch and k. hencken , phys .
c * 56 * , 3054 ( 1997 ) .
j. dobaczewski , w. nazarewicz , t.r .
werner , j.f .
berger , c.r .
chinn , j. decharg , phys .
c 53 ( 1996 ) , 2809 .
a. bulgac , phys .
a * 76 * , 040502(r ) ( 2007 ) .
t. lesinski , t. duguet , k. bennaceur , j. meyer , eur .
j. a * 40 * , 121 ( 2009 ) . | non - empirical effective contact pairing forces to be used in self - consistent mean - field calculations are presented .
these pairing forces , constructed so as to reproduce exactly any given microscopic pairing gaps in infinite homogeneous nuclear matter for any isospin asymmetry , are given in analytical form .
as a by - product , this work provides an analytical solution of the bcs gap equations which could be applied to describe various many - body systems . |
the notion of the intrinsic charm ( ic ) content of the proton has been introduced about 30 years ago in works @xcite .
it was shown that , in the light - cone fock space picture , it is natural to expect a five - quark state contribution of the type @xmath3 to the proton wave function @xcite .
this component can be generated by @xmath4 fluctuations inside the proton where the gluons are coupled to different valence quarks .
the original concept of the charm density in the proton @xcite has nonperturbative nature because the five - quark contribution @xmath3 scales as @xmath5 where @xmath6 is the @xmath7-quark mass @xcite . in the middle of nineties , another point of view on the charm content of the proton
has been proposed in the framework of the variable - flavor - number scheme ( vfns ) of quantum chromodynamics ( qcd ) @xcite .
the vfns is an approach alternative to the traditional fixed - flavor - number scheme ( ffns ) of qcd where only light degrees of freedom ( @xmath8 and @xmath9 ) are considered as active . within the vfns
, the mass logarithms of the type @xmath2 are resummed through the all orders in perturbation theory into a heavy quark density which evolves with @xmath10 according to the standard dglap @xcite evolution equation .
hence this approach introduces the parton distribution functions ( pdfs ) for the heavy quarks and changes the number of active flavors by one unit when a heavy quark threshold is crossed . note also that the charm density arises within the vfns perturbatively via the @xmath11 evolution .
some recent developments concerning the vfns are presented in @xcite .
so , the vfns was introduced to resum the mass logarithms and to improve thus the convergence of perturbative qcd series . at the present time , both nonperturbative and perturbative charm densities are widely used for a phenomenological description of available data .
( a detailed review of the theory and experimental constraints on the charm quark distribution may be found in @xcite ) . in particular , practically all the recent versions of the cteq @xcite and mrst @xcite sets of pdfs are based on the vfn schemes and contain a charm density . at the same time , the key question remains open : how to measure the charm content of the proton ?
the basic theoretical problem is that radiative corrections to the heavy - flavor production cross sections are large : they increase the leading order ( lo ) results by approximately a factor of two .
moreover , soft - gluon resummation of the threshold sudakov logarithms indicates that higher - order contributions can also be substantial ( for reviews , see @xcite . ) on the other hand , perturbative instability leads to a high sensitivity of the theoretical calculations to standard uncertainties in the input qcd parameters : the heavy - quark mass , @xmath6 , factorization and renormalization scales , @xmath12 and @xmath13 , @xmath14 and pdfs .
for this reason , one can only estimate the order of magnitude of the pqcd predictions for charm production cross sections in the entire energy range from the fixed - target experiments @xcite to the rhic collider @xcite .
since production cross sections are not perturbatively stable within qcd , they can not be a good probe of the charm density in the proton .
for this reason , it is of special interest to study those observables that are well - defined in qcd .
nontrivial examples of such observables were proposed in @xcite , where the azimuthal @xmath1 asymmetry and callan - gross ratio @xmath0 in heavy quark leptoproduction were analyzed ( note also the paper @xcite where the perturbative stability of the qcd predictions for the charge asymmetry in @xmath15-quark hadroproduction has been observed ) .
it was shown that , contrary to the production cross sections , the azimuthal asymmetry @xcite and the callan - gross ratio @xcite in heavy flavor leptoproduction are stable within the ffns , both parametrically and perturbatively . in the present paper
, we discuss resummation of the mass logarithms of the type @xmath2 in heavy quark leptoproduction : @xmath16(p_{x } ) .
\label{1}\ ] ] the cross section of the reaction ( [ 1 ] ) may be written as @xmath17,\ ] ] where @xmath18 , while the quantity @xmath19 measures the degree of longitudinal polarization of the virtual photon in the breit frame : @xmath20 @xcite .
the quantities @xmath21 , @xmath22 , and @xmath23 are the usual bjorken kinematic variables while the azimuth @xmath24 is defined in fig .
[ fig1 ] . in next sections
, we will consider resummation of the mass logarithms for the quantities @xmath25 and @xmath26 defined as @xmath27
in our analysis , the acot(@xmath30 ) vfns proposed in @xcite is used . in figs .
[ fig2 ] and [ fig3 ] , we present the lo and next - to - leading order ( nlo ) ffns predictions for the structure function @xmath31 and callan - gross ratio @xmath0 in charm leptoproduction , and compare them with the corresponding acot(@xmath30 ) vfns results . in our calculations ,
the cteq6 m parameterization for pdfs and @xmath32 gev for @xmath7-quark mass is used @xcite .
we convolve the nlo cteq6 m distribution functions with both the lo and nlo partonic cross sections @xcite that makes it possible to estimate directly the degree of stability of the ffns predictions under radiative corrections .
the default common value for the factorization and renormalization scales is @xmath33 .
one can see from fig .
[ fig2 ] that both the radiative corrections and charm - initiated acot(@xmath30 ) contributions to @xmath31 are large : they increase the lo ffns results by approximately a factor of two at @xmath34 for all @xmath23 . at the same time , the relative difference between the nlo ffns and acot(@xmath30 ) predictions in not large : it does not exceed @xmath35 for @xmath36 .
we conclude that it will be very difficult to determine the charm content of the proton using only data on @xmath37 due to large radiative corrections ( with corresponding theoretical uncertainties ) to this quantity .
considering the corresponding predictions for the ratio @xmath25 presented in fig .
[ fig3 ] , we see that in this case the nlo and charm - initiated acot(@xmath30 ) contributions are strongly different .
in particular , the nlo corrections to @xmath25 are small , less than @xmath38 , for @xmath39@xmath40 and @xmath41 .
this implies that large radiative contributions to the structure functions @xmath42 and @xmath43 cancel each other in the ratio @xmath29 with a good accuracy . on the other hand , the charm content of the proton
is predicted to be sizeable within the acot(@xmath30 ) scheme which decreases the lo ffns results by about @xmath44 practically for all values of @xmath45 .
the reason for decreasing of @xmath25 within the vfns is that resummation of the mass logarithms has different effects on the structure functions @xmath46 and @xmath47 .
in fact , contrary to the transverse structure function , @xmath46 , the longitudinal one , @xmath47 , does not contain leading mass logarithms of the type @xmath2 at both lo and nlo @xcite .
for this reason , resummation of the @xmath2 logarithms within the acot(@xmath30 ) scheme leads to increasing of the quantity @xmath46 but does not affect the function @xmath47 .
we conclude that , contrary to the the production cross sections , the callan - gross ratio @xmath0 in deep - inelastic leptoproduction of @xmath7-quark could be a good probe of the charm density in the proton .
note that this observation is practically independent of the pdfs we use .
we have verified that the cteq versions @xcite of the pdfs lead to sizeable reduction of the lo ffns predictions for the ratio @xmath25 . as to the low @xmath48 behavior of the callan - gross ratio , this problem requires resummation of the bfkl @xcite logarithms of the type @xmath49 and will be considered in a forthcoming publication .
fig . [ fig4 ] shows the acot(@xmath30 ) predictions for the azimuthal asymmetry @xmath50 at the following values of variable @xmath51 , @xmath52 , @xmath53 , @xmath54 . for comparison , we plot
also the lo ffns predictions ( solid curves ) .
again , we use the cteq6 m parametrization of pdfs , @xmath55 gev , and @xmath33 .
one can see from fig .
[ fig4 ] the following properties of the azimuthal asymmetry : the mass logarithms resummation leads to a sizeable decreasing of the lo ffns predictions for the @xmath1-asymmetry . within the acot(@xmath30 ) scheme
, the charm - initiated contribution reduces the ffns results for @xmath56 by about @xmath57@xmath58 at @xmath59@xmath40 .
the origin of this reduction is the same as in the case of @xmath25 : contrary to @xmath60 , the azimuth - dependent structure function @xmath61 is safe in the limit @xmath62 at least at lo .
presently , the exact nlo predictions for the azimuth - dependent structure function @xmath61 are not available
. however , in @xcite the nlo corrections to the @xmath1-asymmetry have been estimated within the so - called soft - gluon approximation at @xmath63 .
( note that soft - gluon approximation is unreliable for high @xmath64 ) .
it was demonstrated that soft - gluon corrections to both @xmath61 and @xmath60 are large but cancel each other in their ratio @xmath65 with a good accuracy .
for this reason , it is natural to expect that the azimuthal @xmath1-asymmetry is also a perturbatively stable quantity in wide kinematic range of variables @xmath21 and @xmath10 within the ffns .
we have also analyzed how the vfns predictions for @xmath56 depend on the choice of subtraction prescription .
in particular , the schemes proposed in @xcite have been considered .
we have found that , sufficiently above the production threshold , these subtraction prescriptions also reduce the lo ffns results for the asymmetry by approximately @xmath57@xmath66 . thus , it is shown that impact of the mass logarithms resummation on the @xmath1-asymmetry is essential at @xmath59@xmath40 and therefore can be tested experimentally .
in the present paper , we compare the structure function @xmath60 , callan - gross ratio @xmath67 , and azimuthal asymmetry @xmath68 in charm leptoproduction as probes of the charm content of the proton .
to estimate the charm - initiated contributions , we used the acot(@xmath30 ) vfns @xcite and several cteq sets of pdfs @xcite .
our analysis of the radiative and charm - initiated corrections indicates that , in a wide kinematic range , both contributions to the structure function @xmath31 have similar @xmath21 and @xmath23 behaviors .
for this reason , will be difficult to estimate the charm content of the proton using only data on @xmath31 .
the situation with using the callan - gross ratio and azimuthal asymmetry looks more promising .
our analysis shows that resummation of the mass logarithms leads to reduction of the born predictions for @xmath25 and @xmath26 by @xmath57@xmath66 at @xmath59@xmath40 and @xmath64 .
taking into account the perturbative stability of the callan - gross ratio and azimuthal asymmetry within the ffns established in @xcite , we conclude that the charm density in the proton can , in principle , be determined from high-@xmath23 data on @xmath67 and @xmath68 .
concerning the experimental aspects , the quantities @xmath69 and @xmath56 in charm leptoproduction can be measured in future studies at the proposed eic @xcite and lhec @xcite colliders .
we thank s. i. alekhin and j. bl@xmath70mlein for providing us with fast code @xcite for numerical calculations of the nlo dis cross sections .
the author is grateful to h. avakian , s.j .
brodsky and c. weiss for useful discussions .
this work was supported in part by the ansef 2010 grant ps-2033 and state committee of science of ra , grant 11 - 1c015 .
99 s. j. brodsky , p. hoyer , c. peterson , and n. sakai , phys .
b * 93 * , 451 ( 1980 ) ; s. j. brodsky , c. peterson , and n. sakai , phys .
d * 23 * , 2745 ( 1981 ) .
s. j. brodsky , _
`` light - front qcd '' _ , hep - ph/0412101 ; s. j. brodsky , few body syst . *
36 * , 35 ( 2005 ) .
m. franz , v. polyakov , and k. goeke , phys .
d * 62 * , 074024 ( 2000 ) .
m. a. g. aivazis , j. c. collins , f. i. olness , and w. -k .
tung , phys .
d * 50 * , 3102 ( 1994 ) .
j. c. collins , phys .
d * 58 * , 094002 ( 1998 ) .
v. n. gribov and l. n. lipatov , sov .
. phys . * 15 * , 438 ( 1972 ) ; y. l. dokshitzer , sov .
jetp * 46 * , 641 ( 1977 ) ; g. altarelli and g. parisi , nucl .
b * 126 * , 298 ( 1977 ) .
tung , s. kretzer , and c. schmidt , j. phys .
g * 28 * , 983 ( 2002 ) .
m. kramer , f. i. olness , and d. e. soper , phys .
d * 62 * , 096007 ( 2000 ) .
r. s. thorne , phys .
d * 73 * , 054019 ( 2006 ) ; c. d. white and r. s. thorne , phys .
d * 74 * , 014002 ( 2006 ) ; w. k. tung , h .l .
lai , a. belyaev , j. pumplin , d. stump , and c. -p .
yuan , jhep * 0702 * , 053 ( 2007 ) ; s. kretzer , h. l. lai , f. i. olness and w. -k .
tung , phys .
d * 69 * , 114005 ( 2004 ) ; r. s. thorne and w. k. tung , arxiv:0809.0714 [ hep - ph ] ; p. m. nadolsky , arxiv:0809.0945 [ hep - ph ] ; m. guzzi , p. m. nadolsky , h. l. lai , and c .- p .
yuan , arxiv:1108.5112[hep - ph ] .
j. pumplin , phys .
d * 73 * , 114015 ( 2006 ) ; s. j. brodsky , b. kopeliovich , i. schmidt , and j. soffer , phys .
d * 73 * , 113005 ( 2006 ) ; j. pumplin , h. l. lai , and w. k. tung , phys .
d * 75 * , 054029 ( 2007 ) .
lai , j. huston , s. kuhlmann , f. olness , j. owens , d. soper , w.k .
tung and h. weerts , phys .
d * 55 * , 1280 ( 1997 ) .
lai , j. huston , s. kuhlmann , j. morfin , f. olness , j.f .
owens , j. pumplin and w.k .
tung , eur .
j. c * 12 * , 375 ( 2000 ) .
j. pumplin , d. r. stump , j. huston , h. l. lai , p. nadolsky , and w. k. tung , jhep * 0207 * , 012 ( 2002 ) .
a. d. martin , r. g. roberts , w. j. stirling , and r. s. thorne , phys .
b * 604 * , 61 ( 2004 ) .
e. laenen and s. -o .
moch , phys .
d * 59 * , 034027 ( 1999 ) .
n. kidonakis , phys .
d * 64 * , 014009 ( 2001 ) ; n. kidonakis , phys .
d * 73 * , 034001 ( 2006 ) .
m. l. mangano , p. nason , and g. ridolfi , nucl .
b * 373 * , 295 ( 1992 ) ; s. frixione , m. l. mangano , p. nason , and g. ridolfi , nucl .
b * 412 * , 225 ( 1994 ) .
r. vogt , eur .
j. st * 155 * , 213 ( 2008 ) .
n. ya .
ivanov , a. capella , and a. b. kaidalov , nucl .
b * 586 * , 382 ( 2000 ) .
n. ya .
ivanov , nucl .
b * 615 * , 266 ( 2001 ) .
n. ya .
ivanov , p. e. bosted , k. griffioen , and s. e. rock , nucl .
b * 650 * , 271 ( 2003 ) .
n. ya .
ivanov , nucl .
b * 666 * , 88 ( 2003 ) .
l. n. ananikyan and n. ya .
ivanov , phys .
d * 75 * , 014010 ( 2007 ) .
l. n. ananikyan and n. ya .
ivanov , nucl .
b * 762 * , 256 ( 2007 ) .
n. ya .
ivanov and b. a. kniehl , eur .
j. c * 59 * , 647 ( 2009 ) ; n. ya .
ivanov , nucl .
b * 814 * , 142 ( 2009 ) .
l. g. almeida , g. sterman , and w. vogelsang , phys .
d * 78 * , 014008 ( 2008 ) .
n. dombey , rev .
phys . * 41 * , 236 ( 1969 ) .
s. i. alekhin and j. bluemlein , phys .
b * 594 * , 299 ( 2004 ) .
e. laenen , s. riemersma , j. smith , and w. l. van neerven , nucl .
b * 392 * , 162 ( 1993 ) .
m. buza , y. matiounine , j. smith , r. migneron , and w. l. van neerven , nucl .
b * 472 * , 611 ( 1996 ) .
e. a. kuraev , l. n. lipatov , and v. s. fadin , sov .
jetp * 44 * , 443 ( 1976 ) ; i. i. balitzki and l. n. lipatov , sov .
j. nucl .
* 28 * , 822 ( 1978 ) ; l. n. lipatov , sov .
jetp * 63 * , 904 ( 1986 ) .
s. kretzer and i. schienbein , phys .
d * 58 * , 094035 ( 1998 ) .
a. deshpande , r. milner , r. venugopalan , and w. vogelsang , ann .
nucl . part .
sci . * 55 * , 165 ( 2005 ) .
j. b. dainton , m. klein , p. newman , e. perez , and f. willeke , j. inst . * 1 * , p10001 ( 2006 ) . | we study two experimental ways to measure the heavy - quark content of the proton : using the callan - gross ratio @xmath0 and/or azimuthal @xmath1 asymmetry in deep inelastic lepton - nucleon scattering .
our approach is based on the perturbative stability of the qcd predictions for these two quantities .
we resume the mass logarithms of the type @xmath2 and conclude that heavy - quark densities in the nucleon can , in principle , be determined from data on the callan gross ratio and/or azimuthal asymmetry . in particular ,
the charm content of the proton can be measured in future studies at the proposed large hadron electron ( lhec ) and electron ion ( eic ) colliders . |
in 1986 , the first experimental realisation of a quasi - one - dimensional electron system ( q1des ) in a dynamically confined two - dimensional electron gas ( 2deg ) was achieved @xcite . since
then there has been an extensive theoretical and experimental effort into understanding their basic properties ( see , e.g. , review @xcite and book @xcite and references therein ) .
the interest in these systems stems from two facts .
first , the effective transverse size of a q1des can be easily controlled and made remarkably small , down to the de broglie wavelength of an electron .
this makes it possible to realise experimental systems which have an arbitrary number of occupied transverse modes .
second , the high purity of 2degs grown by molecular beam epitaxy enables the almost collisionless motion of an electron through an experimental q1des .
the coexistence of these two factors has made q1des unique objects for the investigation of transport phenomena yielding , in particular , the observation of the ballistic quantisation of conductance @xcite and the so called 0.7 conductance structure @xcite .
the process of electron transmission through a q1des involves the redistribution of incoming electron flux among its discrete eigenstates followed by adiabatic transport through them .
therefore the determination of the electron eigenstates of a q1des is an integral and very important part of the more general quantum transport problem .
this statement is especially relevant to the ballistic transport regime where the _ total _ conductance of a system is _ completely defined _ by the number of propagating electron modes which in turn can be uniquely calculated from the energy spectrum and the fermi energy @xcite . clearly , the energy spectrum of electrons crucially depends on the effective geometry of a q1des as well as on external and internal fields acting on them . among the possible internal forces ,
one of the least understood examples is the interaction between orbital and spin degrees of freedom of an electron : the _ spin - orbit interaction _ , also referred to as the spin - orbit coupling .
although this interaction has an essentially relativistic nature ( see , e.g. , refs .
@xcite ) , it nevertheless can give rise to an observable modification of semiconductor band structure @xcite . a quite general theoretical approach to the description of the so interaction is to use the following hamiltonian @xcite which stems directly from the quadratic in @xmath0 expansion of the dirac equation : @xmath1 here @xmath2 is the free electron mass , @xmath3 is the momentum operator , @xmath4 is the elementary charge , @xmath5 is the vector of the pauli matrices , @xmath6 is the potential energy of the particle , and @xmath7 stands for the spatial gradient .
the convenience ( or universality ) of the hamiltonian ( [ h_so ] ) is that it does not restrict one to any particular form ( model ) of the potential @xmath6 , but allows freedom of choice depending on the nature and the symmetry of forces present in a given medium .
its form is such that it can remove the spin degeneracy in electron band structure whilst not actually producing an overall spin polarization . in a bulk ( 3d ) crystalline environment ,
the energy @xmath6 arises exclusively from the periodic ( microscopic ) crystal potential .
most multicomponent semiconductors have either zincblende ( gaas and most iii - v compounds ) or wurtzite ( ii - vi compounds ) lattice structure , both of which lack inversion symmetry .
dresselhaus @xcite has shown that this property eventually leads to a so - induced splitting of the conduction band into two subbands .
the magnitude of the splitting is proportional to the cube of the electron wave number @xmath8 . in mosfets and heterostructures the host crystals can not be treated as ideal 3d systems , because the crystal symmetry is broken at the device interface where 2d electron or hole gases are dynamically confined in a quantum well .
the reduction of the effective dimensionality lowers the symmetry of the underlying crystals and results in an additional ( linear in @xmath8 ) term in the dresselhaus splitting . moreover , if the quantum well is sufficiently narrow , then the linear contribution is dominant @xcite and , e.g. , may reach @xmath9
@xmath10 mev in 180- - thick modulation - doped gaas wells @xcite .
theoretical arguments @xcite suggest that this can also be true in strained iii - v crystals and hexagonal ii - vi compounds . along with the _ microscopic _ crystal forces
, there is another source of the potential energy @xmath6 in 2d systems .
it is caused by the interface electric field that accompanies the quantum - well asymmetry @xcite and is directed along the normal to the device plane . since the typical well width is @xmath11
@xmath12 nm , the interface potential turns out to be slowly varying on the scale of the lattice parameter and can be considered _
macroscopic _ , as opposed to the rapidly oscillating atomic field .
the mechanism of the so interaction originating from the interface field was first introduced by rashba in ref .
it also manifests itself as a linear in @xmath8 splitting @xcite of the 2d bandstructure . in a variety of systems including si - mosfets @xcite , inas /
gasb @xcite and alsb / inas / alsb @xcite quantum wells , ingaas / inalas heterostructures @xcite , gaas electron gases @xcite it can be made to dominate the dresselhaus terms indicating the significance of macroscopic potentials in producing observable so effects in low - dimensional systems . in modern nanotechnology there exists a number of methods for creating q1des from 2degs : the split - gate technique ; wet and dry etching ; and cleaved edge overgrowth @xcite .
in essence , they all exploit the confinement of the lateral ( in - plane ) motion of electrons ( holes ) by some transverse potential
. any such potential must be essentially non - uniform in space in order to force the charged particles to remain within a confined area .
the spatial variation scale of the confining potential crucially depends on the particular fabrication method and varies over a wide range : @xmath13 @xmath14 nm .
thus , in q1des one finds another example of macroscopic potentials , viz the lateral confining potential , which is absent in higher - dimensional structures .
the spatial non - uniformity of the confining potential gives rise to an additional ( in - plane ) electric field in the system .
if the confinement is sufficiently strong ( narrow and deep ) , then this field may not be negligible in comparison with the interface - induced ( rashba ) field .
moreover , in nearly square ( i.e. symmetric ) quantum wells where the rashba field is essentially suppressed @xcite , the in - plane ( `` confining '' ) electric field is likely to be dominant .
this suggests the possible importance of the lateral confining potential to the so hamiltonian ( [ h_so ] ) in q1des .
we are not aware of any experimental evidence or measurement of the strength of the so coupling resulting from such a confining potential and can not therefore quote a grounded estimate for the corresponding energy modification .
nevertheless , the above arguments seem sufficiently strong to indicate a possible new mechanism for the so interaction in q1des and point towards new transport effects .
the existence of an additional ( and easily controllable ) source of the so coupling could catalyse experiments on quantum - wire based devices that exploit both the charge and spin of an electron , e.g. spin transistors @xcite and active spin polarisers @xcite .
our belief in the importance of the so interaction in semiconductor q1des is strongly supported by the fact that observable manifestations of a so - induced energy splitting have already been found experimentally @xcite in another type of q1des , viz in electron gases trapped by dislocational defects in silicon crystals .
the source of the so coupling in these systems is an electric field perpendicular to dislocations and the so - related energy splitting is linear in the wave number @xmath8 , which makes a physical analogy between the electron gas on dislocations and semiconductor q1des quite close .
the so interaction manifests itself in spin - dependent electron conductivity along dislocations which can be ascribed to combined spin resonance @xcite of electrons corresponding to transitions between so - split energy levels . in this paper
we analyse theoretically the effect of the spin - orbit interaction on the energy spectrum and conductance of a long q1des . as a model for the q1des
, we use a strictly 2d electron gas subject to a transverse electrostatic confining potential . to decide on a reasonable shape of the confining potential , we assume a sufficiently small effective width ( @xmath15 nm ) for the q1des and a low electron concentration ( @xmath16 @xmath17 ) . combined together , these two factors prevent the confining electric field from being significantly screened by the electron gas . under these conditions
the confining potential can be accurately approximated by a parabolic potential @xcite .
this conclusion is very favourable to our problem because exact analytical expressions for the energy spectrum and wave functions of a 2d electron gas in a parabolic potential are well known and provide us with a good zero approximation in dealing with the so coupling .
we include the so interaction via the hamiltonian ( [ h_so ] ) .
we assume that the potential @xmath6 which is responsible for the so coupling consists of two contributions : ( i ) a parabolic confining potential with a gradient ( or the accompanying electric field ) which lies in the plane of the 2deg ; ( ii ) a potential which arises from the asymmetry of the quantum well with the corresponding electric field ( rashba field ) being uniform and directed perpendicular to the device plane .
we neglect the crystal - field ( dresselhaus ) contribution to @xmath6 .
the goal of this paper is to reveal the qualitative hallmarks of the so interaction in q1des rather than to construct a complete and realistic spectral and transport theory .
therefore we use two major simplifications .
first , we neglect the coulomb interaction between electrons . at first sight
one may not expect this approximation to work in low - concentration 2d electron systems where the coulomb energy may exceed the kinetic energy by an order of magnitude .
however , it has recently been shown theoretically @xcite that the effect of electron - electron interactions on so coupling in such systems can be plausibly taken into account via a renormalisation of the so coupling constant .
more specifically , this renormalisation leads to an enhancement ( by @xmath12
@xmath18 % ) of the strength of the so interaction , which emphasises the significance of the so - related effects in low - dimensional electron systems .
our second simplification is the exploitation of one - band effective mass approximation @xcite for the schrdinger equation . within this approximation ,
the influence of the crystal forces on electron dynamics in the conduction band is reduced to the renormalisation of the electron mass and all interband transitions are left out . despite the obvious simplicity ,
this approach works well @xcite for a wide range of semiconductor materials .
section [ sec - theory ] is the central part of the paper .
it is devoted to the solution of the problem that we have outlined above and to the analysis of the results obtained . in subsection [ subsec-2a ]
we introduce the hamiltonian of a 2deg that includes the parabolic confining potential and the so - interaction term ( [ h_so ] ) . we take into account only those contributions to the so hamiltonian that arise from macroscopic ( i.e. relatively smooth ) potentials , viz from the quantum - well ( rashba ) and the parabolic confining potentials .
since the operator ( [ h_so ] ) contains the pauli matrices , we seek electron wave functions in the form of two - component spinors .
within such a representation the schrdinger equation turns out to be a system of two _ coupled _ differential equations with respect to the spinor components . in next two subsections
we solve this system for two physically different situations .
subsection [ subsec-2b ] deals with the case of zero rashba so coupling , where the entire so interaction comes from the parabolic confining potential .
here the schrdinger equations decouple and the electron wave functions and energy spectrum are found in an explicitly analytical form . in subsection [ subsec-2c ] we consider a more general situation where both mechanisms of the so - coupling are present . in this case
we calculate the electron wave functions and energy spectrum numerically using the results of subsection [ subsec-2b ] as a convenient basis .
afterwards we analyse the energy spectrum obtained in order to reveal the basic features of both so - coupling mechanisms . in subsection [ subsec-2d ]
we discuss possible manifestations of the so interaction in the dependence of the ballistic conductance of a q1des on the fermi energy .
section [ sec - con ] summarises the results of our research .
within one - band effective mass approximation @xcite the hamiltonian of a q1des can be written as @xmath19 where @xmath20 is a 2d position vector and @xmath21 is the effective electron mass . in line with the refs .
@xcite , the _ lateral confining potential _
@xmath22 is approximated by a parabola @xmath23 the quantity @xmath24 controls the strength ( curvature ) of the confining potential .
the in - plane electric field @xmath25 associated with @xmath22 is given by @xmath26 .
we assume that the so interaction hamiltonian @xmath27 ( [ h_so ] ) in eq .
( [ h ] ) is formed by two contributions : @xmath28 .
the first one , @xmath29 , arises from the asymmetry of the quantum well , i.e. from the rashba mechanism @xcite of the so coupling . since the interface - induced ( rashba )
electric field can reasonably be assumed uniform and directed along the @xmath30-axis , the term @xmath29 can be described by the following expression : @xmath31 the so - coupling constant @xmath32 takes values within a range ( @xmath33 @xmath12)@xmath34 ev@xmath35 cm for different systems @xcite . for brevity , in what follows we will refer to the rashba mechanism of the so coupling as @xmath32-coupling .
the second contribution @xmath36 to @xmath37 comes from the parabolic confining potential ( [ v_sg ] ) : @xmath38 here @xmath39 is the typical spatial scale associated with the potential @xmath40 ( [ v_sg ] ) . in eq .
( [ h_b ] ) we have introduced the so - coupling constant @xmath41 .
comparison of typical electric fields originated from the quantum - well and lateral confining potentials allows one to conclude that a plausible estimate for @xmath41 should be roughly @xmath12% of @xmath32 .
moreover , in square quantum wells where the value of @xmath32 is considerably diminished @xcite ( by an order of magnitude ) the constant @xmath41 may well compete with @xmath32 .
henceforth we adopt the term @xmath41-coupling for the mechanism of the so - interaction arising from the lateral confining potential ( [ v_sg ] ) .
our objective is to find eigenvalues and eigenfunctions of the schrdinger equation @xmath42 with the hamiltonian @xmath43 given by eqs .
( [ h ] ) ( [ h_b ] ) .
the wave function @xmath44 is a two - component spinor and the energy @xmath45 is measured from the conduction band edge .
it is easy to see that the hamiltonian ( [ h ] ) ( [ h_b ] ) is translationally invariant in the @xmath46-direction .
we therefore seek solutions to schrdinger s equation in the form of plane waves propagating along the @xmath46-axis , i.e. @xmath47 . here @xmath48 is the dimensionless transverse coordinate , @xmath49 is the transverse wave function , and @xmath50 is the longitudinal wave number . after substituting this representation into the schrdinger equation , we obtain a system of two differential equations with respect to the spinor components @xmath49 : @xmath51 \label{plus } \\
\nonumber \\ & \phi_\downarrow^{\prime\prime}+ \left(\varepsilon_x - t^2-t_\beta\ , t\right ) \phi_\downarrow(t)= \nonumber \\ & ( l_\omega / l_\alpha ) \left[(k_yl_\omega)\phi_\uparrow(t ) - \phi_\uparrow^\prime \right ] , \label{minus}\end{aligned}\ ] ] @xmath52 where @xmath53 is the dimensionless transverse energy , @xmath54 , and the prime denotes a derivative with respect to @xmath55 .
the lengths @xmath56 and @xmath57 defined by @xmath58 are characteristic spatial scales associated with the @xmath32- and @xmath41-couplings , respectively .
we note that the functions @xmath49 depend on three dimensionless external parameters : @xmath59 , @xmath60 , and @xmath61 .
the equations ( [ plus ] ) and ( [ minus ] ) are arranged in such a way that all the terms which couple them together are collected on the rhs. it is interesting that this arrangement separates explicitly the @xmath32 and @xmath41 mechanisms of the so interaction .
indeed , the @xmath32-terms enter only the rhs , while all the @xmath41-terms are contained on the lhs. this suggests that the @xmath41-coupling is responsible for forming independent ( `` unperturbed '' or `` non - interacting '' ) wave functions @xmath49 , while the @xmath32-coupling mixes them together to form the solution of the whole system ( [ plus ] ) and ( [ minus ] ) .
we start the analysis of eqs .
( [ plus ] ) and ( [ minus ] ) with the situation where the quantum well is square so that the interface - induced electric field vanishes . in terms of the so - interaction
this means that the rashba mechanism ( i.e. the @xmath32-mechanism ) of the so - coupling can be omitted .
in other words , the characteristic length @xmath56 of the @xmath32-interaction tends to infinity : @xmath62 , or @xmath63 .
the study of this relatively simple case will provide us with a convenient basis for treating the general problem with finite @xmath56 . once @xmath59 has been put equal to zero , the rhs in eqs .
( [ plus ] ) and ( [ minus ] ) vanish and they transform into two independent hermite equations @xcite whose exact solutions are @xmath64 } { \sqrt{2^n n!\pi^{1/2 } } } h_n\left(t \mp t_\beta\right ) \nonumber \\ & & \qquad \qquad \qquad \qquad n=0,1,2,\ldots \label{phi0}\end{aligned}\ ] ] here @xmath65 is the hermite polynomial of @xmath66-th order . the real wave functions @xmath67 and @xmath68 form complete sets with respect to the discrete quantum number @xmath66 and are normalised by the following conditions : @xmath69 where both @xmath70 and @xmath71 are taken at a same value of @xmath50 . from eq .
( [ phi0 ] ) one can see that the both `` up '' ( @xmath72 ) and `` down '' ( @xmath73 ) spinor components have exactly the same shape as functions of @xmath55 but are spatially displaced with respect to each other by an amount of @xmath74 ( see fig .
[ fig1 ] ) . according to the definition ( [ tb ] ) ,
this displacement is a direct consequence of the @xmath41-coupling because it does not vanish as long as @xmath75 ( except where @xmath76 ) . at finite values of @xmath74 , the complete sets of functions @xmath67 and @xmath68 turn out _ not _ to be mutually orthogonal , i.e. @xmath77 . instead
, the scalar cross - product @xmath78 in the case of sufficiently weak @xmath41-coupling ( @xmath79 ) is governed by the following asymptotics : @xmath80 @xmath81 @xmath82 @xmath83 the `` displacing effect '' of the @xmath41-coupling on spinor wave functions [ see eq . [ phi0 ] and fig .
[ fig1 ] ] is superficially similar to the effect of a perpendicular magnetic field on a q1des .
however , the essential difference is that a magnetic field shifts _ both _ spinor components @xmath84 and @xmath85 _ as a whole _ by an amount proportional to the strength of the magnetic field ( see , e.g. , ref .
@xcite ) , but it does _ not _ affect their _ mutual _ spatial distribution .
in contrast to this , as can be seen in fig .
[ fig1 ] , the @xmath41-coupling causes a spatial separation of the spin - polarised electron states @xmath84 and @xmath85 . although the @xmath41-coupling shifts the spinor components @xmath86 apart , it nevertheless is not capable of lifting their energy degeneracy .
the dimensionless transverse energy @xmath87 corresponding to the both eigenfunctions @xmath86 ( [ phi0 ] ) is given by ( see fig . [ fig2 ] ) @xmath88 the total electron energy @xmath45 then forms parabolic subbands for each @xmath66-th transverse mode : @xmath89 for zero @xmath41-coupling ( @xmath90 ) formulas ( [ spec0 ] ) and ( [ e ] ) describe the well - known electric subbands @xcite .
we now examine the situation where the effective geometry of the quantum well is such that the interface - induced electric field is non - zero .
an example is the triangular well which forms at a heterojunction interface @xcite .
this interface - induced field gives rise to a finite @xmath32-coupling and therefore to finite values of the coupling constant @xmath59 .
for arbitrary non - zero values of @xmath59 , eigenvalues of eqs .
( [ plus ] ) and ( [ minus ] ) can not be obtained as simply as they were in the case of @xmath91 . to analyse the energy spectrum in the presence of @xmath32-coupling
, we will follow the scheme proposed in ref .
namely , we first expand the unknown functions @xmath49 in terms of the unperturbed solutions @xmath92 ( [ phi0 ] ) of eqs .
( [ plus ] ) and ( [ minus ] ) : @xmath93 @xmath94 we then substitute the expansions ( [ exp1 ] ) and ( [ exp2 ] ) into eqs .
( [ plus ] ) and ( [ minus ] ) respectively and make use of the property @xcite @xmath95 to remove the derivative with respect to @xmath55 on the rhs of eqs .
( [ plus ] ) and ( [ minus ] ) .
we next multiply the equations obtained by @xmath67 and @xmath68 respectively and integrate them over @xmath55 from @xmath96 to @xmath97 . owing to the orthogonality condition ( [ norm ] ) , the summation over @xmath98 is removed by the delta - function @xmath99 and we find the simple vector relations : @xmath100 @xmath101 here @xmath102 and @xmath103 are vectorised coefficients of the expansions ( [ exp1 ] ) and ( [ exp2 ] ) respectively , and @xmath104 are tridiagonal matrices defined by their elements : @xmath105 @xmath106 @xmath107 note that the matrices @xmath104 are neither symmetric nor anti - symmetric .
( [ rel1 ] ) ( [ u - lower ] ) illustrate the effectiveness of the representations ( [ exp1 ] ) and ( [ exp2 ] ) . indeed , by using the functions @xmath108 as expansion bases
, we have reduced differential operators on the lhs of eqs .
( [ plus ] ) and ( [ minus ] ) to the scalar factor @xmath109 . in other words ,
matrices which were supposed to act on @xmath110 and @xmath111 in eqs .
( [ rel1 ] ) and ( [ rel2 ] ) turn out to be diagonal within the representations ( [ exp1 ] ) and ( [ exp2 ] ) .
our next step is to find the relationship between the vectors @xmath112 and @xmath113 . to do this , we equate the rhs of eq .
( [ exp1 ] ) to that of eq .
( [ exp2 ] ) and take the scalar product of the resulting equation with @xmath67 . after making use of the orthogonalisation condition ( [ norm ] ) , we find that @xmath114 where the matrix @xmath115 is defined by @xmath116 according to eq .
( [ phi0 ] ) [ see also eqs .
( [ nn ] ) ( [ nn+p ] ) ] the matrices @xmath117 are _ not _ diagonal as long as the @xmath41-coupling is _
analogously , by taking a scalar product of the rhs of eqs .
( [ exp1 ] ) and ( [ exp2 ] ) by @xmath68 , it can be shown that @xmath118 finally , we combine relations ( [ rel1 ] ) , ( [ rel2 ] ) , ( [ rel3 ] ) , and ( [ rel4 ] ) into a closed homogeneous equation with respect to @xmath110 : @xmath119 . in order for this equation to have a non - trivial solution , the jacobian matrix must satisfy the following condition : @xmath120 the roots @xmath121 of this equation determine the dispersion law of electrons ( see , e.g. , ref .
@xcite ) . here , @xmath122 as @xmath63 .
alternatively , the relations ( [ rel1 ] ) , ( [ rel2 ] ) , ( [ rel3 ] ) , and ( [ rel4 ] ) could have been resolved with respect to @xmath111 , which would have led to the permuted jacobian matrix @xmath123 .
however , such a permutation leaves the determinant unchanged and therefore the dispersion equation ( [ det ] ) would have been still applicable . to obtain the solution of eq .
( [ det ] ) for given values of the external parameters @xmath59 , @xmath60 , and @xmath61 we truncated the infinite matrix @xmath124 down to the first @xmath125 rows and columns and used a numerical root finder to obtain the zeros of its determinant . owing to the conveniently chosen bases
( [ exp1 ] ) and ( [ exp2 ] ) , the roots of the equation ( [ det ] ) converge very quickly to their exact values as @xmath125 is increased .
for instance , for @xmath59 , @xmath126 , it suffices to take @xmath127 to find the ten lowest energy levels with a net error @xmath128 .
the energy spectrum @xmath129 calculated by this procedure is shown in fig .
let us first consider the case of negligibly weak @xmath41-coupling , when @xmath130 [ fig .
[ fig3](a ) ] . here
we see two - fold spin degeneracy of all quantum levels at @xmath131 .
as soon as we move away from the point @xmath131 , the so interaction lifts the degeneracy and produces an energy splitting @xmath132 proportional to @xmath50 , where @xmath133 are energies associated with the `` up '' and `` down '' spin polarisations respectively .
this linear behaviour agrees with both theoretical predictions @xcite and experimental observations @xcite of the rashba splitting in @xmath134 systems .
the linear dependence of @xmath135 on @xmath50 can be easily deduced analytically from eqs .
( [ plus ] ) and ( [ minus ] ) within first - order perturbation theory with respect to the parameter @xmath59 .
indeed , if we treat the rhs of eqs .
( [ plus ] ) and ( [ minus ] ) as perturbations , then the first - order correction @xmath136 to the unperturbed energy ( [ spec0 ] ) will be given by @xmath137 or @xmath138 here @xmath139 are diagonal matrix elements of perturbation operators @xmath140 : @xmath141 @xmath142 to calculate the matrix elements @xmath139 we use the property ( [ deriv - prop ] ) and take into account that @xmath143 [ see eqs .
( [ nn ] ) ( [ nn+p ] ) ] . as a result
, we obtain @xmath144 and hence the rashba splitting @xmath135 ( [ dr0 ] ) is governed by @xmath145 i.e. it is indeed proportional to the longitudinal wave number @xmath50 . the perturbative result ( [ dr1 ] ) applies as long as @xmath146 , i.e. if @xmath147 . for @xmath148
this condition restricts @xmath61 to values much less than about @xmath149 . in fig .
[ fig3](a ) we see that the linear behaviour of the energy splitting for all the curves holds true until @xmath150 ( dotted line )
. within the region @xmath151 the asymptotic ( [ dr1 ] ) gives a very good fit to all the energy branches .
however , as soon as @xmath61 becomes larger than @xmath152 , the dispersion curves start to bend .
eventually this leads to an _ anticrossing _ of branches corresponding to quantum levels with neighbouring discrete numbers @xmath66 .
this fact drastically contrasts the simpler situation with no confining potential ( [ v_sg ] ) , where the linearly split spectrum ( [ dr1 ] ) represents the exact solution @xcite .
usually @xcite anticrossing is associated with similar symmetry ( `` hybridisation '' ) of underlying states ( wave functions ) . to reveal possible symmetries in our case , we return to the perturbation theory with respect to @xmath59 and consider first - order corrections @xmath153 to the wave functions ( [ phi0 ] ) : @xmath154 owing to the orthogonality condition ( [ norm2 ] ) the second ( constant ) term in eq .
( [ p0 ] ) does _ not _ contribute to @xmath153 ( [ dphi ] ) for any @xmath155 .
at the same time , according to eqs .
( [ deriv - prop ] ) and ( [ norm2 ] ) , the first ( differential ) term in eq .
( [ p0 ] ) gives rise to _ non - zero _ matrix elements @xmath156 for @xmath157 and @xmath158 . as a result ,
@xmath153 ( [ dphi ] ) takes the form : @xmath159 the total ( perturbed ) wave function is given by @xmath160 .
[ fig4 ] shows graphs of @xmath161 for @xmath162 and @xmath163 .
( [ dphi0 ] ) demonstrates that the @xmath66-th quantum state is no longer independent of its `` neighbours '' , i.e. of the @xmath164-st states . in turn
, the @xmath165-st state now depends on its own nearest neighbours , i.e. on both the @xmath66-th and the @xmath166-nd states .
owing to this interstate coupling , all spinors @xmath167 partially acquire the symmetry properties of spinors @xmath168 .
moreover , since the result ( [ dphi0 ] ) is perturbative , it is plausible that the interstate coupling becomes more pronounced when the perturbative approach breaks down , i.e. when @xmath169 .
this `` hybridisation '' of electron states accounts for the anticrossing of neighbouring energy branches in fig .
[ fig3](a ) . after passing the relatively narrow anticrossing region @xmath170 , all the curves in fig .
[ fig3](a ) straighten out and adopt a very accurate linear behaviour for @xmath171 .
there are two main features that attract attention in the infinite interval @xmath171 : ( i ) there are no further anticrossings of approaching energy branches . even those with closest quantum numbers @xmath66 , which anticrossed at smaller values of @xmath61 , now cross ; ( ii ) all the straight lines @xmath172 go down ( up ) with the same slope @xmath173 which is independent of a level number @xmath66 .
although statement ( i ) means asymptotical smallness ( rather than absolute absence ) of anticrossings in the region of large @xmath171 in fig .
[ fig3](a ) , these anticrossings are much weaker than any other anticrossings in the system and therefore can be left out . to explain these properties we should return to eqs .
( [ plus ] ) , ( [ minus ] ) , and ( [ phi0 ] ) .
according to eq .
( [ phi0 ] ) , the spatial variation scale of functions @xmath86 is of the order unity .
this means that the derivatives @xmath174 on the rhs of eqs .
( [ plus ] ) and ( [ minus ] ) respectively can be roughly estimated as @xmath175 , at least for @xmath176 and sufficiently small values of @xmath59 .
hence , it seems plausible that within the region , where @xmath177 , the terms @xmath174 can be neglected in comparison with @xmath178 . taking this into account together with the condition @xmath90 , we reduce eqs .
( [ plus ] ) and ( [ minus ] ) to @xmath179 @xmath180 unlike the general problem ( [ plus ] ) and ( [ minus ] ) , eqs .
( [ plus1 ] ) and ( [ minus1 ] ) can be simply decoupled by a unitary transformation in the spin space .
the corresponding matrix has the form @xmath181 it is straightforward to verify that eigenvalues ( energies ) of the decoupled equations are @xmath182 and eigenfunctions are identical to those given by eq .
( [ phi0 ] ) at @xmath90 .
the formula ( [ eps - large ] ) yields the expected linear dependence of both @xmath183 and @xmath184 on @xmath61 with slopes @xmath185 independent of a level number @xmath66 .
the asymptotics ( [ eps - large ] ) give a very accurate fit to the spectral pattern in the region @xmath171 in fig .
[ fig3](a ) . since eqs .
( [ plus1 ] ) and ( [ minus1 ] ) can be decoupled by a simple rotation , states @xmath186 and @xmath187 turn out to be intrinsically independent of each other as well as of all the other states .
this fact alone explains why the anticrossing of energy branches is not observed for sufficiently large values of @xmath61 .
let us now consider how the energy spectrum is modified by switching on the @xmath41-coupling [ fig .
[ fig3](b ) ] . from visual comparison of figs .
[ fig3](a ) and ( b ) we see that the main effect of the @xmath41-coupling is to _ enhance _ considerably the anticrossing of `` neighbouring '' spectrum branches .
moreover , the strength of the anticrossing now depends on the quantum number @xmath66 and grows as @xmath66 is increased .
an interesting consequence of this behaviour is an _ essential reduction _ of the linear rashba energy splitting @xmath188 , contrasting the expectation that the additional mechanism of the so interaction should intensify the splitting rather than suppress it . to understand the peculiarities of fig .
[ fig3](b ) we note that in subsection [ subsec-2b ] it was shown that a finite @xmath41-coupling leads to spatial separation between `` up '' and `` down '' spinor components ( see fig . [ fig1 ] ) . as a result
, the orthogonality condition ( [ norm2 ] ) no longer applies .
instead , the asymptotics ( [ nn ] ) ( [ nn+p ] ) should be used .
if we now calculate the rashba splitting @xmath135 ( [ dr0 ] ) , then we will see that scalar products @xmath189 give non - zero ( viz linear in @xmath60 ) contributions to the diagonal matrix elements @xmath139 , so that @xmath135 is now described by [ cf .
( [ dr1 ] ) ] @xmath190 \frac{l_\omega}{l_\alpha}(k_yl_\omega ) .
\label{dr2}\ ] ] from this formula it immediately follows that the energy splitting @xmath135 _ diminishes _ in comparison with fig .
[ fig3](a ) by an amount @xmath191 that is _ proportional _ to the quantum level number @xmath66 .
this conclusion agrees well with data presented in fig .
[ fig3](b ) . the result ( [ dr2 ] ) can be interpreted in the language of interstate coupling .
actually , it is easy to verify that the asymptotics ( [ nn ] ) ( [ nn+p ] ) give rise to an additional ( proportional to @xmath60 ) term @xmath192 in the wave function correction @xmath153 ( [ dphi0 ] ) : @xmath193 \label{dphib}\end{aligned}\ ] ] this term , as well as @xmath194 ( [ dphi0 ] ) , involves `` nearest '' unperturbed wave functions @xmath195 . since parameters @xmath59 and @xmath60 are independent of each other ,
we conclude that the @xmath41-coupling enhances the hybridisation between an @xmath66-th and @xmath164-st electron states and thereby the anticrossing of corresponding energy branches .
the strength of the hybridisation grows with the growth of @xmath66 . because of the enhanced interstate coupling , the anticrossing of neighbouring energy branches in fig .
[ fig3](b ) is not restricted to a narrow region of relatively small values of @xmath196 as it was in fig .
[ fig3](a ) . instead , we can see that for all @xmath197 there exists a second anticrossing in the region @xmath198 .
moreover , for larger quantum numbers @xmath199 a third anticrossing emerges at @xmath200 .
to apply the results of the previous subsections to the study of the effect of the so interaction on the ballistic conductance of a long q1des at low temperature , we must relate its conductance to its energy spectrum .
here we do this using the two - probe landauer formula @xcite @xmath201 where @xmath202 is the ballistic conductance , @xmath203 is the fermi energy , and @xmath204 is the number of occupied electron subbands which propagate in the same direction : @xmath205 . \label{m}\ ] ] here @xmath206 is the energy of the @xmath207-th minimum in the @xmath66-th electron subband with the spin orientation @xmath208 .
@xmath209 is the heaviside unit step function .
as will be seen later the so interaction can produce multiple minima in each 1d subband owing to the anticrossings between different subbands .
since @xmath206 can be found directly from the dispersion law of electrons , the conductance @xmath202 ( [ g ] ) turns out to be _ completely defined _ by the energy spectrum alone .
therefore , as long as the expression ( [ g ] ) applies , the knowledge of the electron energy levels in a q1des will be sufficient to predict the behaviour of the ballistic conductance as a function of the fermi energy . the applicability of eq .
( [ g ] ) , as well as of the general scattering approach @xcite to quantum transport , is essentially based on the condition that a current must travel in any 1d electron subband without scattering into any other . in ref .
@xcite it was shown that this _ current conservation _ condition holds true in quite general circumstances including placing a q1des in both a finite external magnetic field and an arbitrary external electrostatic potential . however , the proof in ref .
@xcite ignored spin degrees of freedom and therefore did not take into account potentials acting in the spin space .
such neglect is not valid for our problem where the so interaction hamiltonian ( [ h_so ] ) leads to a highly non - trivial role for the spin in forming the energy spectrum of electrons ( see subsections [ subsec-2a ] [ subsec-2c ] ) .
for that reason we can not rely on the conclusion of ref .
@xcite but should check explicitly if the current is still conserved in the presence of a finite so coupling .
the first step is to define the matrix elements of the current density @xmath210 for the case where the wave functions @xmath211 are spinors @xcite , @xmath212 : @xmath213^\dagger\hat{\bbox{\sigma}}\psi_n\biggr\}. \label{jmn}\ ] ] here the dagger denotes the hermitian conjugate and @xmath214 .
it is straightforward to verify that the divergence of the vector @xmath210 is given by @xmath215 we now suppose that the hamiltonian of the system has the general form @xmath216 , where @xmath217 is a hermitian operator ( @xmath218 ) that acts in both coordinate and spin spaces .
as applied to our problem , @xmath219 [ see eq .
( [ h ] ) ] . using this hamiltonian and the schrdinger equation @xmath42 , we express @xmath220 and @xmath221 in terms of @xmath222 and @xmath223 respectively and substitute the expressions obtained into eq .
( [ dj ] ) . as a result
, we have @xmath224 where @xmath225 and @xmath226 are energies corresponding to the states @xmath227 and @xmath223 respectively . in the absence of any inelastic collisions
, any scattering occurs between states of the same energy .
so , without loss of generality , we can restrict ourselves to considering only equal energies @xmath228 in eq .
( [ dj2 ] ) , in which case we find that @xmath229 this fundamental identity ensures _ local _ current conservation in the system .
once it has been established , if the system is translationally invariant in the longitudinal direction , further steps in proving the _ global _ conservation of current do not depend on specifics of the hamiltonian and can be carried out in line with ref .
. therefore we arrive at the conclusion that the total current ( i.e. current integrated over the cross section of the channel ) between states @xmath98 and @xmath66 at @xmath228 is equal to zero unless @xmath230 . in other words , eigenstates of the hamiltonian @xmath43
are perfect current - carrying states that are free from scattering even in the presence of arbitrary so coupling . for our problem
this result implies that the spectrum of the hamiltonian ( [ h ] ) is directly relevant to and completely defines the ballistic conductance in the presence of the so coupling .
this allows us to use the simple landauer formula ( [ g ] ) , ( [ m ] ) , in which the minima @xmath206 of the energy subbands can be found from the analysis of the spectrum presented in subsections [ subsec-2a ] [ subsec-2c ] .
now we are in a position to discuss the features of the ballistic conductance in a q1des subject to the so interaction . for illustrative purposes we start with the `` ideal '' case of zero so coupling .
the corresponding subband energies @xmath231 are plotted in fig .
[ fig5](a ) ( solid curves ) as functions of @xmath61 .
the dependence @xmath232 can simply be deduced from this figure by moving a horizontal line @xmath233 from zero upwards and counting the number of points at which this line crosses the spectral parabolas . since all the subbands in an ideal system are two - fold spin degenerate for any @xmath50 , this number coincides with the number @xmath21 of propagating modes in the q1des . as a result , we restore the well - known picture @xcite of ballistic conductance quantisation with equidistant jumps each of height @xmath234 [ solid curve in fig .
[ fig6](a ) ] .
dotted curves in fig .
[ fig5](a ) show subband energies @xmath235 in the presence of the @xmath41-coupling [ see eq .
( [ spec0 ] ) ] .
each subband is two - fold spin degenerate .
it is seen that for any given value of @xmath203 the number @xmath21 of forward propagating modes remains the same as it was in the ideal limit .
thus , we conclude that the @xmath41-coupling alone does _ not _ affect the ballistic conductance . to detect the presence of the @xmath41-coupling in a system one should exploit a unique feature that can only be due to the @xmath41-coupling . as a reasonable example of such a feature
, we mention the spatial separation between spinor components that was established in subsection [ subsec-2b ] and illustrated in fig .
[ fig1 ] . owing to this effect
, the regions of the most probable distribution of electrons with opposite spin orientations will be shifted apart in the transverse direction . by analogy with ferromagnetic materials ,
this effect can manifest itself in the appearance of the magnetisation of the electron system and its dependence on the transverse coordinate .
now we move on to the more complex case of finite @xmath32-coupling . to start with
let us imagine the fictitious situation where the linear rashba energy splitting @xmath236 [ see eq .
( [ dr1 ] ) and fig .
[ fig3](a ) ] holds true not only for small values of @xmath61 but for all @xmath61 . here
, the subband energies would be given by @xmath237 .
this formula describes parabolas of the same form as the solid curves in fig .
[ fig5](a ) but shifted by an amount @xmath238 along the abscissa and lowered vertically by @xmath239 .
the parabolas @xmath240 were used in ref .
@xcite to predict spontaneous spin polarisation of ballistic electrons in quantum wires due to spin splitting .
whereas the _ real _ energy branches @xmath241 and @xmath242 anticross [ see fig . [ fig3](a ) ] , the fictitious dispersion curves @xmath243 and @xmath244 _ cross _ at the point @xmath245 .
this point remains the same for all numbers @xmath66 and goes to infinity as the so coupling vanishes .
it is easy to verify that for @xmath246 the crossing point @xmath245 lies to the right of the point @xmath247 at which the parabola @xmath244 has the minimum . at the same time , if @xmath248 , then the crossing point turns out to be to the left of the minimum point . below
we will see that these simple conclusions following from the _ fictitious _ energy spectrum play an important role in determining the behaviour of the conductance @xmath232 .
[ fig5](b ) presents the energy subbands @xmath249 ( solid and dashed curves ) of a q1des in the case of _ weak _ @xmath32-coupling when @xmath246 ( more specifically , @xmath250 ) .
dotted lines indicate the _ fictitious _ energy levels @xmath251 and @xmath252 . in full accordance with the above conclusions , these levels cross to the _ right _ of the bottom of the parabola @xmath252 . as a result
, the crossing angle turns out to be quite small . from fig .
[ fig5](b ) it is seen that this angle determines essentially the shape of the _ real _ energy curves @xmath253 and @xmath254 within the anticrossing region @xmath255 .
as long as the angle is small , the anticrossing remains very smooth and the curves @xmath253 and @xmath254 behave very much like the ideal parabolas in fig . [
fig5](a ) .
the same observation is also true for higher quantum numbers @xmath66 .
such similarity suggests that weak @xmath32-coupling ( @xmath256 ) does not have a strong effect on the conductance .
indeed , it is easy to see by scanning fig .
[ fig5](b ) with the horizontal line @xmath233 that the only effect of weak @xmath32-coupling on @xmath232 is to shift the conductance quantisation steps down to lower energies by an amount @xmath239 [ see dashed lines in fig .
[ fig6](a ) ] .
as soon as the coupling constant @xmath59 gets over the threshold of @xmath257 , the spectral picture becomes much more interesting .
[ fig5](c ) shows a case when @xmath258 . here
the parabolas @xmath251 and @xmath252 ( dotted curves ) cross to the _ left _ of the bottom of @xmath244 , which makes the crossing angle relatively large .
the direct consequence of this is that the energy branches @xmath253 and @xmath254 can not now anticross as smoothly as they did for @xmath256 .
instead , in order to keep continuity , the lower curve [ i.e. @xmath253 ] is forced to exhibit a non - monotonic portion ( `` bump '' ) within the anticrossing region .
although the height of the bump is rather small ( @xmath260 in the dimensionless units or @xmath261 mev as determined from typical subband spacings @xcite ) , its geometry is a unique feature of the so coupling mechanism [ see a magnified plot of the bump in fig . [ fig5](d ) ] .
the remarkable fact about it is that the bump contains a region where the electron energy decreases as the wavenumber grows . in other words
, there exists an energy interval within which electrons are allowed to have a negative longitudinal group velocity @xmath262 for positive @xmath50 .
since this interval has a finite width and is surrounded by energy domains with positive group velocities , the curvature of the line @xmath263 reverses sign within the negative - velocity interval and therefore there exists an inflection point where @xmath264 [ see fig . [ fig5](d ) ] . at this point
the effective electron mass @xmath265 diverges .
since the current passing through a 1d subband is proportional to the product of the group velocity and the density of states @xcite , the singular effective mass has no effect on the conductance in the absence of electron scattering .
however , it could significantly change the dependence @xmath232 in the presence of a scattering mechanism ( e.g. , disorder or electron - electron interaction ) .
as far as transport through 1d electron subbands is concerned , it is clear from fig .
[ fig5](d ) that the negative propagating mode coexists with a forward propagating mode with the same spatial wave function . in this physical situation
it is likely that weak elastic scattering between these states would result in directed localisation ref .
@xcite so that they would not be observed in conductance .
however , if both forward and backward electron modes contribute equally to the conductance , then this immediately brings us to the conclusion that the bump on the curve @xmath253 gives rise to a peak in the dependence @xmath232 [ see fig . [
fig6](b ) ] .
the height of the peak is @xmath234 and its width is defined by the height of the bump .
analogous bumps can be observed for all quantum levels @xmath266 as a consequence , the sharp peak is seen in fig .
[ fig6](b ) not only on the first conductance plateau @xmath267 but also on the second one @xmath268 and would still be seen on all further plateaus . of course , the strong @xmath32-coupling ( @xmath248 ) , as well as the weak coupling ( @xmath248 ) , shifts the conductance quantisation steps to lower values of the fermi energy in comparison with the ideal situation where @xmath91 . however
, in contrast to the weak - coupling limit , strong coupling makes this shift much larger for the first step than for the others .
this effect is clearly seen in fig .
[ fig6](b ) as opposed to the ideal conductance quantisation shown by dotted lines .
the larger shift of the first step is explained by the fact that the @xmath162 quantum state does not have a neighbour with the next lowest quantum number @xmath66 .
the energy level @xmath269 is therefore not forced to anticross with any other ( lower lying ) energy branch and therefore nothing affects its linear ( rashba ) dependence on @xmath50 [ see fig . [ fig3](a ) ] .
the final step of our analysis is to consider the case where both strong @xmath32-coupling ( @xmath258 ) and relatively weak @xmath41-coupling ( @xmath270 ) are present in the system .
the corresponding energy bands are shown in fig .
[ fig5](e ) . as we demonstrated in subsection [ subsec-2c ] , finite @xmath41-coupling tends to suppress the energy splitting caused by the @xmath32-coupling [ cf .
[ fig3](a ) and [ fig3](b ) and see eq .
( [ dr2 ] ) ] .
effectively , this suppression can be interpreted as the enhancement of the anticrossing between neighbouring energy branches @xmath241 and @xmath242 . in fig .
[ fig5](e ) this effect is seen as a decrease in the gap between energy branches @xmath241 and @xmath271 ( @xmath272 ) in comparison with fig .
[ fig5](c ) .
moreover , in accordance with eq .
( [ dr2 ] ) , the gap becomes monotonically smaller as @xmath66 grows .
the enhanced anticrossing has a drastic effect on the energy bumps created by the strong @xmath32-coupling : all the bumps except for the lowest one are now smoothed away and do not give rise to the two additional ( forward and backward ) propagating modes . as a result , the conductance exhibits only one sharp peak that occurs on the plateau @xmath267 [ see fig . [
fig6](c ) ] .
a second effect of the enhanced anticrossing on the conductance is that the conductance quantisation steps ( starting from the second one ) are now located closer to the ideal steps than they were for zero @xmath41-coupling .
as each next step appears , its distance from the corresponding ideal step diminishes until eventually they fuse .
there is another effect of the enhanced energy anticrossing on the conductance that is not seen in figs .
[ fig6 ] but could possibly be detected in the presence of impurities .
it is found experimentally @xcite that the ballistic conductance quantisation breaks down as the constriction becomes too long ( at least @xmath273 @xmath274 m ) .
this effect can be attributed to potential fluctuations caused by the random distributions of remote impurities .
@xcite showed analytically that the degradation of quantised conductance decreases exponentially with the ratio of the 1d subband spacings to the standard deviation of impurity potential fluctuations .
obviously , the subband spacings depend on the presence of the so interaction ( see figs . [ fig5 ] ) . by comparing figs .
[ fig5](c ) and ( e ) we see that a finite @xmath41-coupling pushes neighbouring energy branches apart and thus increases the subband spacings ( possibly by tens of per cent ) .
this means that the presence of the @xmath41-mechanism of the so interaction could allow the experimental observation of the conductance quantisation structure in longer samples .
we have studied theoretically the influence of the spin - orbit ( so ) interaction on the energy spectrum and the ballistic conductance of a quasi-1d electron system ( q1des ) formed by lateral electric confinement of a 2d electron gas .
the presence of the confining potential proves to be crucial in two ways .
first , it alters significantly the effect of the quantum well asymmetry on the energy spectrum in comparison with a purely 2d system . as it was shown by rashba _
@xcite , in 2d electron gases this asymmetry gives rise to so coupling ( which we refer to as the @xmath32-coupling ) that manifests itself in a lifting of the spin degeneracy of electronic states .
the accompanying energy splitting @xmath135 was found to be proportional to the in - plane electron wave number @xmath8 .
we have demonstrated that in a q1des the function @xmath275 is _ non - monotonic _ [ fig .
[ fig3](a ) ] with the standard linear dependence @xcite @xmath276 being observed for relatively small and large values of @xmath8 only .
such a drastic change in behaviour is explained by the essentially different effect of so coupling operator @xmath277 ( [ p0 ] ) on unperturbed wave functions depending on whether or not the lateral confining potential is taken into account .
indeed , in the zero - potential case the unperturbed wave functions are plane waves characterised by the continuous in - plane wave vector @xmath278 and the spin .
the action of the operator @xmath279 on any such wave function reduces to a simple renormalisation of its amplitude with the renormalising factor being independent of coordinates . as a consequence of this simple effect ,
wave functions belonging to different values of @xmath278 remain independent of each other and the so coupling manifests itself in linear ( monotonic ) energy splitting @xmath276 which is the same for all quantum states .
in contrast to this , the wave functions in the presence of a finite confining potential have a more complicated structure ( [ phi0 ] ) and are characterised by the discrete quantum number @xmath66 instead of the continuous transverse wave number . according to eq .
( [ deriv - prop ] ) , the action of the operator @xmath277 on the @xmath66-th wave function includes projecting it onto states with the next ( preceding ) closest quantum numbers @xmath280 . in turn
, the @xmath164-st states are projected onto their `` neighbours '' with numbers @xmath66 and @xmath281 respectively . as a result , the operator @xmath277 _ couples _ effectively any @xmath66-th and @xmath164-st wave functions . in other words , once the so coupling has been taken into account , the @xmath66-th and @xmath282-st wave functions cease to be independent and possess symmetry elements of each other .
this partial symmetry between states leads to an _ anticrossing _ of the closest ( neighbouring ) energy branches in fig .
[ fig3](a ) and hence to the non - monotonic dependence @xmath275 .
apart from the interplay with the familiar ( quantum - well - asymmetry or rashba ) mechanism of the so interaction , the lateral confining potential by itself appears to be a source of additional dynamical coupling between the orbital and spin degrees of freedom of an electron .
this coupling ( which we refer to as @xmath41-coupling ) originates from the natural spatial non - uniformness of the confining potential .
a typical variation scale of this potential lies within the wide range @xmath13
@xmath14 nm , which makes the accompanying electric field sufficiently strong to compete with the quantum - well - asymmetry field .
this competition may become especially noticeable in square quantum wells with a relatively weak rashba contribution . whereas the quantum - well field is normal to the device plane ,
the confinement - induced electric field is parallel to the plane .
it is also spatially non - uniform as long as the coordinate dependence of the confining potential is more complex than linear .
these features make the @xmath41-coupling an essentially different ( new ) mechanism of the so interaction in a q1des which _ can not _ be taken into account by simply adjusting the rashba interaction constant .
this claim is confirmed by fig .
[ fig3](b ) that demonstrates the combined effect of both the rashba and the confinement - induced couplings on the energy spectrum . from comparison of fig .
[ fig3](b ) with fig . [ fig3](a )
we see that the major role of @xmath41-coupling is to _ reduce _ the rashba energy splitting @xmath135 ( see region @xmath151 ) rather than to give a positive correction to it .
it is crucial that the reduction depends on the quantum number @xmath66 and monotonically grows as @xmath66 increases .
this fact is a clear manifestation of the position dependence of the electric field created by the confining potential and indicates the independent nature of @xmath41-coupling .
the suppression of the rashba energy splitting is a part of the overall effect of the @xmath41-coupling on the underlying electron system which consists of a considerable enhancement of the hybridisation between different quantum states and eventually leads to a more pronounced anticrossing of neighbouring energy branches .
in addition to the investigation of the so - effects in the energy spectrum of electrons , we have discussed the possible manifestations of the so interaction in the ballistic conductance of a q1des .
the key point of our approach to the conductance is the fundamental _ current - conservation identity _ that was proven in subsection [ subsec-2d ] . according to this identity
, the electron eigenstates that were found as the solution to the spectral problem are _ perfect _ current - carrying states .
a current can travel in any of these states without scattering into any other .
this property therefore allows the ballistic conductance to be calculated directly from the energy spectrum with the help of simple landauer formula ( [ g ] ) .
an analysis of the ballistic conductance @xmath202 reveals that the @xmath41-coupling alone does _ not _ affect the dependence of @xmath202 on the fermi energy @xmath203 .
this fact is illustrated by fig . [ fig5](a ) where we see that the @xmath41-coupling reduces the curvature of the parabolic energy bands ( cf . dotted and solid curves ) , while the band edges ( bottoms of the parabolas ) remain anchored .
thus , @xmath41-coupling can not be detected by measuring the ballistic conductance in systems where the so interaction appears _ only _ in a form of the @xmath41-coupling ( e.g. in square wells ) .
instead , one should use experimental methods that allow direct observation of the electron energy spectrum in a q1des , e.g. magnetotunneling measurements @xcite . in the presence of @xmath32-coupling the behaviour of @xmath232
is essentially determined by the strength of the so interaction .
if the @xmath32-coupling is not too strong ( @xmath256 ) , then its only effect on the conductance will be shifts of the conductance quantisation steps to lower fermi energies in comparison with an ideal ( i.e. with zero so interaction ) situation .
we note that such shifts are not related to the lateral confining potential and should also be present in purely 2d systems .
this effect should be detectable in transconductance measurements ( see , e.g. , ref .
@xcite ) which determine both the conductance and the subband spacings simultaneously . in the limit of strong @xmath32-coupling ( @xmath248 )
the electron energy bands take on a very interesting form [ fig .
[ fig5](c ) ] .
the most remarkable feature is the appearance of narrow energy intervals where two additional ( forward and backward propagating ) electron modes exist [ fig .
[ fig5](d ) ] .
such intervals can be found in each 1d subband , starting from the lowest one .
the additional electron modes have similar magnitude group velocities but propagate in opposite directions .
they have almost identical subband wave functions and therefore would be susceptible to strong intermode scattering in the presence of disorder .
however , in a sufficiently pure q1des , the additional electron modes give rise to the unusual periodic sharp steps in @xmath232 shown in fig .
[ fig6](b ) .
this picture is changed by switching on relatively weak ( @xmath283 ) @xmath41-coupling . as we mentioned above in the discussion of the energy spectrum ,
the @xmath41-coupling enhances the anticrossing of energy levels initiated by the @xmath32-coupling .
as applied to the conductance , this effect leads to _ quenching _ the sharp conductance peaks by the @xmath41-coupling .
the existence of the single peak ( or just a few of peaks ) in the dependence of @xmath202 on @xmath203 could be a clear experimental indication of the presence of the @xmath41-coupling in the system .
the most crucial point for the experimental observation of the conductance peaks in fig .
[ fig6 ] is to make the ratio @xmath59 sufficiently large ( @xmath284 ) . in typical systems ,
where enhancing the so interaction is not paid special attention , the value of @xmath59 hardly exceeds @xmath285 .
an additional , at least threefold , increase in @xmath59 could be achieved by using : ( i ) materials with light carrier masses ; ( ii ) strong ( narrow ) lateral potential confinement @xmath286 nm ; ( iii ) heterojunctions ( triangular well ) rather than quantum - well heterostructures ( square well ) ; ( iv ) a back gate voltage to maximise the interface ( rashba ) electric field .
we hope that the results presented in this paper will stimulate further experimental and theoretical work with the aim of understanding the role of the spin - orbit interaction in determining the transport properties of quasi-1d systems .
a. yacoby , h. l. stormer , k. w. baldwin , l. n. pfeiffer , and k. w. west , solid state commun . * 101 * , 77 ( 1997 ) ; a. yacoby , h. l. stormer , n. s. wingreen , l. n. pfeiffer , k. w. baldwin , and k. w. west , , 4612 ( 1996 ) .
v. v. kveder , yu .
a. osipyan , and a. i. shalynin , pisma zh .
. fiz . * 40 * , 10 ( 1984 ) [ jetp lett . * 40 * , 10 ( 1984 ) ] ; v. v. kveder , v. ya .
kravchenko , t. r. mchedlidze , yu .
a. osipyan , d. e. khmelnitski , and a. i. shalynin , pisma zh .
fiz . * 43 * , 202 ( 1986 ) [ jetp lett . * 43 * , 25 ( 1986 ) ]
. r. landauer , ibm journ .
development * 1 * , 223 ( 1957 ) ; r. landauer , philos . mag . *
21 * , 863 ( 1970 ) ; r. landauer , in _ localization , interaction , and transport phenomena _ edited by g. bergmann and y. bruynseraede ( springer - verlag , new york , 1985 ) . g. timp , r. behringer , s. sampere , j. e. cunningham , and r. e. howard , in _ proccedings of the international symposium on nanostructure physics and fabrication _ , edited by m. a. reed and w. p. kirk ( academic , new york , 1989 ) , p. 331 | we discuss the effect of the spin - orbit interaction on the band structure , wave functions and low temperature conductance of long quasi - one - dimensional electron systems patterned in two - dimensional electron gases ( 2deg ) .
our model for these systems consists of a linear ( rashba ) potential confinement in the direction perpendicular to the 2deg and a parabolic confinement transverse to the 2deg .
we find that these two terms can significantly affect the band structure introducing a wave vector dependence to subband energies , producing additional subband minima and inducing anticrossings between subbands .
we discuss the origin of these effects in the symmetries of the subband wave functions . 2 |
the @xmath0 semifilled shell mn( ... @xmath0@xmath1 ,
@xmath2 ) atom has long served as the bridge to , and touchstone for , a better understanding of the interaction of transition metal atoms with x - ray and vacuum - ultraviolet radiations from early days ( see , e.g. , works by connerade et.al .
@xcite , davis and feldkamp @xcite , amusia et.al .
@xcite ) to now ( see review papers by sonntag and zimmermann @xcite , martins et.al .
@xcite , as well as some of the most recent papers by frolov et.al .
@xcite , osawa et.al . @xcite , hirsch et.al . @xcite and references therein ) . the structure and spectra of the mn atom
are of self - interest as well , in view of the found abundance of unique features associated with its semifilled @xmath0 subshell .
in contrast , studies of another process of the basic and applied significance - electron scattering off the mn atom - are too scarce .
the mn atom presents a special interest for studying electron scattering processes .
this is because it belongs to a class of atoms with the highest spin multiplicity .
the latter is owing to co - directed spins of all five electrons in the mn @xmath0 subshell , due to hund s rule . as of today , understanding of electron scattering off mn is rudimentary .
the only available experimental data relate to corresponding e@xmath8mn differential scattering cross sections ( dsc ) measured at a _ single _ value of the electron energy @xmath14 ev by williams et.al .
@xcite and meintrup et.al .
the same stands for theoretical studies as well .
thus , to understand and interpret experiment , amusia and dolmatov @xcite calculated the @xmath15 ev e@xmath8mn elastic dcs in the framework of a multielectron simplified `` spin - polarized '' random phase approximation with exchange ( sprpae ) @xcite , meintrup et.al .
@xcite did it in a r - matrix framework , and recently remeta and kelemen @xcite performed their calculations of said dcs in the framework of a local spin density approximation , but , again , only at discrete values of the electron energies of @xmath16 and @xmath15 ev .
whatever interesting effects were found in the above cited works they provide only a limited insight into a problem , since things might work quite differently at other , especially lower electron energies but just @xmath16 and @xmath15 ev energies .
clearly , studying the scattering process through a continuum spectrum of electron energies is a way toward a deeper understanding of , as well as discovering new trends in , e@xmath8mn elastic scattering , in particular , and e@xmath8@xmath17 , in general .
however , no such study has been performed to date , to the best of these authors knowledge .
its performance is , thus , not merely a wish but necessity .
it is the ultimate aim of the present paper to study e@xmath8mn elastic scattering through a continuum spectrum of electron energies , to fill in an important gap in the current knowledge on said process . to meet this end
, we focus on calculations of e@xmath8mn elastic scattering phase shifts @xmath4 and corresponding total cross section @xmath6 in the electron energy range of @xmath18 ev . the performed calculations , following the work @xcite , utilize a concept of the reducible self - energy part @xmath19 of the green function @xmath20 of an incoming electron . the calculations are carried out in three consequentially growing levels of sophistication .
first , this is a one - electron `` spin - polarized '' hartree - fock ( sphf ) approximation @xcite which is the zero - order approximation in perturbation theory in the coulomb interelectron interaction @xmath21 for the green function . in refs .
@xcite , sphf was adapted , as well as proven to be applicable , to the description of elastic electron scattering off semifilled shell atoms .
second , @xmath19 is calculated in the second - order approximation in perturbation theory in @xmath21 , to account for electron correlation in the system .
this approximation is known as a simplified random phase approximation with exchange @xcite and , with sphf being chosen as the zero - order approximation with respect to electron correlation interactions , is referred to as sprpae@xmath22 , in the present paper .
third , for a fuller account of electron correlation in electron scattering , @xmath19 is determined beyond sprpae@xmath22 approximation by solving the dyson equation for @xmath19 of a scattered electron , as in ref .
such approximation is referred to as sprpae@xmath23 in the paper . by comparing results obtained in the framework of each of the sphf , sprpae@xmath22 , and sprpae@xmath23 approximations
we unravel several important trends in e@xmath8mn scattering ; these have already been noted in abstract and are subject to a detailed discussion in text .
atomic units ( a.u . ) are used throughout the paper unless specified otherwise .
a convenient starting approximation to study the structure and spectra of semifilled shell atoms is a `` spin - polarized '' hartree - fock ( sphf ) approximation @xcite . over the years
, sphf has been extensively and successfully exploited by the authors of this paper and their colleagues ( see , e.g. , refs .
@xcite ) by using it directly , or utilizing it as the zero - order approximation for a multielectron random phase approximation with exchange ( rpae ) @xcite , to study photoionization of said atoms and their ions . as noted earlier , sphf
was also used successfully by amusia and dolmatov @xcite to provide the initial understanding and interpretation of experimental data of williams et.al .
@xcite on the @xmath15 ev e@xmath8mn elastic dcs .
recently , sphf has been employed by remeta and kelemen @xcite to adapt a local density approximation to electron scattering off semifilled shell atoms , including mn , to provide a further understanding of results of experiments @xcite .
applicability of sphf , as well as the generalized on its basis other methods @xcite , for calculations and understanding of said phenomena with semifilled shell atoms - participants is out of doubt .
the quintessence of sphf is as follows .
it accounts for the fact that spins of all electrons in a semifilled subshell of the atom ( e.g. , in the @xmath0 subshell of mn ) are co - directed , in accordance with hund s rule .
it will be assumed in the present paper that spins of said electrons point upward ( @xmath9 ) .
this results in splitting of each of other closed @xmath24 subshells in the atom into two semifilled subshells of opposite spin orientations , @xmath25@xmath9 and @xmath25@xmath10 .
this is in view of the presence of exchange interaction between @xmath26@xmath9 electrons with spin - up electrons in the original semifilled subshell of the atom but absence of such for @xmath26@xmath10 electrons .
thus , the atom of concern of this paper - the mn atom - has the following sphf configuration : mn( ... @xmath27@xmath9@xmath27@xmath10@xmath28@xmath9@xmath29@xmath9@xmath30@xmath10 , @xmath31s ) .
sphf equations for the ground or excited states of a semifilled shell atom differ from ordinary hf equations for closed shell atoms ( see , e.g. @xcite ) by accounting for exchange interaction only between electrons with the same spin orientation ( @xmath9 , @xmath9 or @xmath10 , @xmath10 ) . by solving sphf equations , one determines radial parts @xmath32 and @xmath33 of the wavefunctions of spin - up and spin - down electrons in the ground , excited , or scattering state of the atom . for the continuum energy spectrum of scattered electrons ,
@xmath34 have the well - known for a central field asymptotic behavior at large @xmath35 : @xmath36 here , @xmath37 , @xmath38 , @xmath39 , and @xmath40 are the momentum , orbital momentum , energy , and the phase shift of a scattered electron , respectively . the total electron elastic scattering cross sections of spin - up ( @xmath41 ) and spin - down ( @xmath42 ) electrons are determined as @xmath43 a simplified random phase approximation with exchange , version-@xmath22 ( sprpae@xmath22 ) , accounts for electron correlation in a e@xmath44 system in the second - order perturbation theory in the coulomb interelectron interaction @xmath21 between the incoming and atomic electrons @xcite .
the approximation exploits the concept of the reducible self - energy part of the one - electron green function @xmath19 of a spin - up , @xmath45 , or spin - down , @xmath46 , scattered electron . in the framework of sprpae@xmath22 ,
@xmath47 is illustrated with the help of feynman diagrams in fig .
[ fig1 ] . of the green function of a scattered electron as defined in sprpae@xmath22 .
here , a line with a right arrow denotes an electron , whether a scattered electron ( lines marked by @xmath48 , @xmath49 ) or an atomic excited electron ( a line @xmath50 ) , a line with a left arrow denotes a vacancy ( hole ) in the atom ( lines @xmath51 and @xmath52 ) , a wavy line denotes the coulomb interelectron interaction @xmath21 . on the other hand ,
the notations @xmath48 , @xmath49 , @xmath50 , @xmath51 , and @xmath52 themselves stand for corresponding electronic states : @xmath53@xmath54 , @xmath55@xmath56 , _
etc_.,width=302 ] in fig .
[ fig1 ] , diagrams ( a ) and ( c ) are called `` _ _ direct _ _ '' diagrams , in contrast to `` _ _ exchange _ _ '' diagrams ( b ) and ( d ) .
the latter two are due to exchange interaction in a e@xmath44 system . basically , the diagrams in fig .
[ fig1 ] illustrate how an incoming electron `` @xmath57 '' polarizes a @xmath51-subshell in the atom by causing actual or virtual excitations @xmath51@xmath58@xmath50 from the subshell , and couples with these excited states itself via the coulomb interaction .
note , exchange diagrams ( b ) and ( d ) in fig .
[ fig1 ] vanish whenever spin directions of an incoming electron and polarized subshell of the atom are opposite , due to orthogonality of electron spin - functions .
we do not account for spin - flip effects since they are negligible in e@xmath8mn scattering @xcite , at the energies of interest .
once @xmath59 is calculated , the elastic electron scattering phase shifts of spin - up and spin - down electrons are determined as @xcite @xmath60 here
, @xmath61 is the correlation correction term to the sphf calculated phase shift @xmath62 : @xmath63 a mathematical expression for the matrix element @xmath64 is obtained with the help of the many - body correspondence rules @xcite ; we refer the reader for details to ref .
@xcite .
when the energy of a scattered electron exceeds the ionization threshold of the atom - scatterer , the term @xmath61 and , thus , the phase shift @xmath65 itself become complex @xcite .
correspondingly , @xmath66 here , @xmath67 and @xmath68 are the real and imaginary parts of @xmath65 , respectively : @xmath69 the total electron elastic scattering cross section is then determined @xcite by @xmath70 in the context of the present paper , @xmath71 , @xmath72 , and @xmath73 .
a version-@xmath23 of the simplified random phase approximation with exchange ( sprpae@xmath23 ) provides a fuller account for electron correlation .
there , the reducible self - energy part of the one - electron green function of a scattered electron is sought as the solution of corresponding dyson equation @xcite .
the latter , in an operator form , in terms of spin - up and spin - down electrons , is @xmath74 here , @xmath75 and @xmath76 are the operators of the irreducible and reducible self - energy components of the green - function operator for an incoming electron , respectively , and @xmath77 is the operator of the green function in the framework of sphf : @xmath78 , where @xmath79 is the sphf hamiltonian operator of the system . in sprpae@xmath23 , to avoid tremendous calculation difficulties , the general dyson equation , eq .
( [ eq9 ] ) , is simplified @xcite .
this is achieved by replacing the operator of the irreducible self - energy component of the green function @xmath75 by @xmath80 ( see fig . [ fig1 ] ) , to a good approximation . as in sprpae@xmath22 , corresponding sprpae@xmath23 phase shifts @xmath65 and total cross sections @xmath81 are determined by eqs .
( [ eq3])-([eq8 ] ) .
in the present paper , e@xmath8mn elastic electron scattering is investigated in the electron energy range between approximately @xmath82 and @xmath83 ev , as a case study .
this is because core - polarization effects in e@xmath8mn elastic electron scattering are expected to be particularly strong at low electron energies .
a trial calculation showed that , at the given energies , accounting for contributions of only @xmath84 , @xmath85 , @xmath86 , and @xmath87 partial waves to the total elastic scattering cross section , as well as accounting for only monopole , dipole , quadrupole , and octupole excitations of the atomic core in calculations of @xmath76 is an excellent approximation .
sphf , sprpae@xmath22 , and sprpae@xmath23 calculated data for real ( @xmath88 ) and imaginary ( @xmath89 ) parts of elastic scattering phase shifts @xmath90 of spin - down electronic waves are depicted in figs .
[ fig2 ] and [ fig3 ] , and those for @xmath91 and @xmath92 of @xmath93 for spin - up electronic waves - in figs .
[ fig4 ] and [ fig5 ] .
( dotted line ) and sprpae@xmath23 ( solid line ) calculated data for real ( @xmath88 ) and imaginary ( @xmath89 ) parts of the e@xmath8mn elastic scattering phase shifts @xmath94 of @xmath84 and @xmath85 spin - down electronic waves versus the electron energy @xmath39.,width=264 ] ( dotted line ) and sprpae@xmath23 ( solid line ) calculated data for real ( @xmath88 ) and imaginary ( @xmath89 ) parts of the e@xmath8mn elastic scattering phase shifts @xmath90 of @xmath86 and @xmath87 spin - down electronic waves.,width=264 ] ( dotted line ) and sprpae@xmath23 ( solid line ) calculated data for real ( @xmath91 ) and imaginary ( @xmath92 ) parts of the e@xmath8mn elastic scattering phase shifts @xmath95 of @xmath84 and @xmath85 spin - up electronic waves.,width=264 ] ( dotted line ) and sprpae@xmath23 ( solid line ) calculated data for real ( @xmath91 ) and imaginary ( @xmath92 ) parts of the e@xmath8mn elastic scattering phase shifts @xmath95 of @xmath86 and @xmath87 spin - up electronic waves.,width=264 ] the depicted data unravel important trends in low energy e@xmath8mn elastic scattering which are detailed below .
first , one can see from figs .
[ fig2]-[fig5 ] that both sprpae@xmath22 and sprpae@xmath23 correlation affects strongly all phase shifts quantitatively and in a number of cases - for @xmath85 , @xmath86 , and @xmath87 partial waves - even qualitatively , compared to sphf calculated data .
thus , the utter importance of electron correlation in the e@xmath8mn low energy elastic electron scattering is revealed .
second , notice how sprpae@xmath23 calculated data for low energy @xmath96 phase shifts differ drastically from calculated data obtained in the framework of sprpae@xmath22 .
indeed , the sprpae@xmath23 calculated phase shift @xmath97 ( fig .
[ fig3 ] ) drops abruptly , with decreasing electron energy , from @xmath98 to @xmath99 rad between @xmath100 to @xmath101 ev , then develops a maximum at yet lower energies , after which it approaches its final value of @xmath102 at @xmath103 . clearly , the described behavior of sprpae@xmath23 calculated @xmath97 has little in common with that obtained in the framework of sprpae@xmath22 .
strong quantitative differences take place between sprpae@xmath22 and srpar@xmath23 calculated data for @xmath104 as well , see fig .
[ fig5 ] . also , notice abrupt changes in sprpae@xmath23 calculated phase shifts @xmath105 between approximately @xmath82 and @xmath23 ev , in opposition to those calculated in sprpae@xmath22 .
the unraveled drastic differences between sprpae@xmath23 and sprpae@xmath22 calculated phase shifts is a novel result ; it was not observed in previous similar calculations performed for other atoms .
thus , the present work reveals , and generally proves , the necessity for a fuller account ( as in sprpae@xmath23 ) of correlation beyond the second - order approximation ( sprpae@xmath22 ) for an adequate understanding of e@xmath8mn scattering .
third , which is of significant importance , notice differences between phases of scattered electronic waves with the same @xmath38s but opposite spin orientations , @xmath106 versus @xmath93 .
said differences are drastic for @xmath107 as well as @xmath108 spin - up and spin - down waves in all three sphf , sprpae@xmath22 , and sprpae@xmath23 approximations ( cp .
[ fig3 ] and [ fig5 ] ) .
for other spin - up and spin - down waves the differences in question are less spectacular , but , nevertheless , exist as well .
thus , it is revealed in the present study that , generally , scattering of oppositely spin - polarized electrons off the mn atom take different routes .
below , we provide reasons for the unraveled trends in the e@xmath8mn electron scattering phase shifts . in sphf ,
the dependence of e@xmath8mn scattering phase shifts on a spin - orientation of a scattered electron is a straightforward effect .
it is due to the presence / absence of exchange interaction between , respectively , @xmath109@xmath9/@xmath109@xmath10 incoming electrons and primarily five @xmath110@xmath9 electrons in the semifilled @xmath0@xmath9 subshell in the atom .
this is a characteristic feature of elastic electron scattering off any semifilled shell atom , for an obvious reason .
next , since no stable negative mn ion exists with the ground - state atomic core configuration @xcite , the number @xmath111 of bound @xmath86@xmath9 states in the atom is @xmath112 ( due to the presence of @xmath0@xmath113 subshell in the atom ) , whereas corresponding @xmath114 .
consequently , in accordance with the generalized levinson s theorem [ @xmath115 as @xmath116 , @xmath117 being the number of bound states with given @xmath38 in the field of an atom and @xmath118 being the number of occupied @xmath119 states in the atom itself ] @xcite , @xmath120 , whereas @xmath121 as @xmath116 .
this translates into the initial drastic quantitative and qualitative differences between sphf calculated phase shifts @xmath122 ( fig . [ fig3 ] , dashed line ) and @xmath123 ( fig .
[ fig5 ] , dashed line ) . in sprae@xmath22 , compared to sphf , differences between @xmath124 and @xmath125 scattering phases become even more spectacular .
indeed , one can see ( fig .
[ fig3 ] , dotted line ) that the sprpae@xmath22 calculated real part @xmath126 of @xmath124 is a monotonic function of @xmath39 in the interval of @xmath100 to @xmath82 ev , but the real part @xmath127 of @xmath125 ( fig .
[ fig5 ] , dotted line ) is not .
the latter now has a well developed minimum followed by an appreciable maximum with decreasing energy @xmath128 ev .
this results , partly , in an additional , compared to sphf , quantitative as well as qualitative difference between @xmath124 and @xmath125 , due to the effects of electron correlation .
next , when sprpae@xmath22 correlation is accounted for , no lesser spectacular differences also emerge between the calculated @xmath129 and @xmath130 scattering phases .
indeed , sprpae@xmath22 correlation is seen to induce strong differences between the real parts @xmath131 ( fig .
[ fig3 ] , dotted line ) and @xmath132 ( fig .
[ fig5 ] , dotted line ) of phase shifts @xmath129 and @xmath130 : @xmath133 turns into an oscillating function of @xmath39 whereas @xmath134 does not .
a trial calculation showed that the sprpae@xmath22 correlation induced differences between @xmath135 and @xmath136 are due primarily to polarization of the @xmath0@xmath9 subshell by an incoming @xmath109 electron .
thus , when the incoming electron is a spin - up electron , exchange diagrams ( b ) and ( d ) with @xmath137@xmath9@xmath138 in fig .
[ fig1 ] contribute to phase shifts , whereas their contributions vanish for a spin - down incoming electron , as was explained earlier in text .
higher order sprpae@xmath23 correlation corrections , compared to sprpae@xmath22 , induce additional significant changes , primarily in @xmath139@xmath140 and @xmath139@xmath141 scattering phases . while the impact of said corrections on a real part @xmath142@xmath143 of @xmath139@xmath141 ( fig .
[ fig5 ] , solid line ) results mostly in its quantitative change compared to corresponding sprpae@xmath22 data , it ( the impact ) makes a real part @xmath142@xmath144 of @xmath139@xmath140 to become both quantitatively and qualitatively different than @xmath142@xmath145 , fig .
[ fig3 ] , solid line .
the unraveled different impact of sprpae@xmath23 correlation on @xmath139@xmath140 compared to @xmath139@xmath141 is because the sprpae@xmath23 equation for the reducible self - energy part of the green - function , eq .
( [ eq9 ] ) , contains many cross - product terms between terms associated with each of the diagrams depicted in fig .
note , there are no cross - products terms in the framework of sprpae@xmath22 at all . since the number of @xmath110 electrons in the mn atom is a @xmath146 unbalanced in favor of @xmath110@xmath9 electrons , many sprpae@xmath23 cross - product terms , involving various excitations of @xmath110@xmath9 electrons , vanish for an incoming spin - down electron but remain for a spin - up scattered electron .
this aggravates the difference between the sprpae@xmath23 correlation impact on scattering of spin - up compared to spin - down electrons off the atom compared to the difference emerging in sprpae@xmath22 calculations . for @xmath107 incoming electrons ,
this difference is huge , as was illustrated above .
thus , the carried out study reveals that the lower - order sprpae@xmath22 approximation is clearly insufficient for a proper understanding of electron scattering off a semifilled shell atom .
this is an important finding .
note , the demonstrated fast variation of the sprpae@xmath23 @xmath124 phase with energy in a quite narrow energy region near @xmath100 ev can be interpreted as a prominent time - delay of the partial @xmath107@xmath10 wave while crossing the atomic region .
indeed , long ago , wigner @xcite connected the time of quantum mechanical reaction procedure @xmath147 with the derivative on the phase over energy . in the considered case of up and down electrons
, one has @xmath148 with the help of this expression , one can estimate not only the time duration of the scattering process of a given @xmath109 wave , but also the time duration difference @xmath149 for up and down partial waves @xmath38 .
for the @xmath107@xmath9 and @xmath107@xmath10 waves in @xmath150mn scattering near @xmath12 ev , we find that @xmath151 @xmath152 s , in the framework of sprpae@xmath23
. to get insight into both the individual correlation and exchange interaction impacts , as well as their combined effect , on _ observable _ elastic electron scattering characteristics , we calculated the e@xmath8mn total spin - up @xmath41 and spin - down @xmath42 elastic scattering cross sections in the framework of sphf , sprpae@xmath22 , and sprpae@xmath23 .
the performed calculations utilized the above presented data for phase shifts .
the thus calculated @xmath41 and @xmath42 are depicted in fig .
[ fig6 ] .
( dotted line ) and sprpae@xmath23 ( solid line ) calculated data for the total elastic scattering cross section @xmath13 of spin - up electrons off the mn atom .
middle panel : the same as in upper panel but for @xmath11 of spin - down electrons .
bottom panel : individual data for @xmath13 and @xmath11 calculated in the framework of sprpae@xmath23 ( for a clearer comparison of these two final results ) .
inset : the total spin - averaged elastic scattering cross section @xmath153 , see eq .
( [ eq10 ] ) , calculated in the framework of sprpae@xmath23.,width=264 ] the depicted results are self - illustrating in the demonstration of the found significance of correlation impacts on e@xmath8mn elastic electron scattering cross sections of spin - up and spin - down electrons .
clearly , a fuller account of correlation , secured by the sprpae@xmath23 approximation , is seen to be decisive
. of significant interest is unraveling of the existence of a narrow resonance in @xmath42 near @xmath154 ev in the framework of sprpae@xmath23 ( see the middle and bottom panels in fig . [ fig6 ] ) .
this resonance has an interesting nature which is associated both with a semifilled shell structure of , and electron correlation in , the mn atom . indeed , as was discussed and demonstrated above , fig .
[ fig3 ] , it is because of the semifilled shell structure of the mn atom that the elastic scattering phase shift @xmath155 as well as real parts @xmath126 of both @xmath156 and @xmath157 phase shifts drop to a zero at @xmath158 . on the way to a zero ,
@xmath124 as well as @xmath126 pass through the value of @xmath159 , at certain values of the electron energy .
correspondingly , at such energy , a resonance emerges in a partial @xmath160 elastic scattering cross section both in sphf , where @xmath161 , eq .
( [ eq2 ] ) , and sprpae@xmath22 or sprpae@xmath23 , where @xmath160 depends on @xmath162 , eq .
( [ eq8 ] ) .
naturally , the existence of this resonance in partial @xmath160 translates into its emergence in the total elastic scattering cross @xmath42 section at the same electron energy as well . indeed , as seen in fig .
[ fig6 ] , the resonance in question is present in @xmath163 and @xmath164 at @xmath12 ev , as well as in @xmath165 at @xmath166 ev . the resonance is broad and weakly developed both in the sphf and sprpae@xmath22 calculated cross sections . in contrast , it is sharp and well developed in @xmath164 .
sprpae@xmath23 , without doubt , is a more complete approximation than either of the sphf or sprpae@xmath22 approximations .
prediction based on the basis of sprpae@xmath23
should , thus , match reality better than those in the framework of sphf or sprpae@xmath22 . the sprpae@xmath23 calculated results in question ,
thus , allow us to claim the discovery of the _ actual _ existence of the @xmath100 ev _ sharp _ resonance in the e@xmath8mn total elastic electron scattering cross section .
furthermore , as follows from the discussion , the resonance in question can exist neither without the semifilled shell nature of the mn atom nor without accounting for electron correlation in the e@xmath8mn scattering process .
this permits us to speak about the discovery of a novel type of a resonance in electron - atom scattering which we name the _ semifilled - shell - correlation _ resonance , to stress its uniqueness .
note , the semifilled - shell - correlation resonance shows up not only in the total elastic electron scattering cross section @xmath42 of spin - down electrons but in corresponding spin - averaged total cross section @xmath167 as well , see inset in bottom panel of fig .
the latter - @xmath167 - was calculated as @xmath168 where @xmath169 @xmath170 being the spin of the atom ( @xmath171 for the mn atom ) and @xmath172 being the electron spin .
in summary , it is unraveled in the present paper that ( a ) sprpae@xmath22 and sprpae@xmath23 electron correlation affects significantly e@xmath8mn electron elastic scattering phase shifts and cross sections , ( b ) correlation may affect , and in the case of the mn atom does affect , drastically differently scattering phase shifts of electrons with opposite spin - orientations , ( c ) a fuller account of correlation ( as in sprpae@xmath23 ) beyond the second - order approximation of perturbation theory in the coulomb interelectron interaction ( sprpae@xmath22 ) may be , and in the case of e@xmath8mn scattering is , crucial for an adequate understanding of the spectrum of corresponding phase shifts and cross sections versus the electron energy , and ( d ) a combined impact of the semifilled shell nature of the mn atom and electron correlation results in the emergence of a novel type of the resonance - the semifilled - shell - correlation resonance - in the e@xmath8mn total elastic scattering cross section .
we urge experimentalists and other theorists to verify the made predictions .
furthermore , the unraveled physics behind the discovered effects in e@xmath8mn scattering is so clear that the present authors have little doubts that most of the effects are inherent features of electron elastic scattering off any multilectron semifilled shell atom in general , not just off the mn atom .
the predicted effects should even be stronger for atoms - scatterers with a higher spin multiplicity ( than that of mn ) , such as the cr( ... @xmath173@xmath9@xmath174@xmath9 , @xmath175s ) or eu( ... @xmath176@xmath9@xmath177@xmath9@xmath178@xmath10 , @xmath179s ) atoms , or similar .
we are currently undertaking such study .
v.k.d . acknowledges the support of nsf grants no .
phy-@xmath180 and phy-@xmath181 .
connerade , m. w. d. mansfield , and m. a. p. martin , proc .
. lond . * 350 * , 405 ( 1976 ) .
l. c. davis and l. a. feldkamp , phys .
a * 17 * , 2012 ( 1978 ) .
amusia , v. k. dolmatov , and v. k. ivanov , zh .
. fiz . * 85 * , 115 ( 1983 ) [ sov .
jetp * 58 * , 67 ( 1983 ) ] .
b. sonntag and p. zimmermann , rep .
. phys . * 55 * , 911 ( 1992 ) .
m. martins , k. godanhusen , t. richter , p. wernet , and p. zimmermann , j. phys .
b * 39 * , r79 ( 2006 ) .
m. v. frolov , n. l. manakov , a. f. starace , phys .
a * 82 * , 023424 ( 2010 ) .
t. osawa k. kawajiri , n. suzuki , t. nagata , y. azuma , and f. koike , j. phys .
b * 45 * , 225204 ( 2012 ) .
k. hirsch , v. zamudio - bayer , f. ameseder , a. langenberg , j. rittmann , m. vogel , t. mller , b.v .
issendorff , and j. t. lau , phys .
a * 85 * , 062501 ( 2012 ) .
w. williams , j. c. cheesborough , and s. trajmar , j. phys .
b * 11 * , 2031 ( 1978 ) .
r. meintrup , g. f. hanne , and k. bartschat , j. phys .
b * 33 * , l289 ( 2000 ) .
amusia and v. k. dolmatov , zh .
fiz . * 97 * , 1129 ( 1990 ) [ sov .
jetp * 70 * , 632 ( 1990 ) ] .
amusia and l. v. chernysheva , _ comput . atomic processes _
( iop publishing ltd . ,
bristol , 1997 ) .
e. yu . remeta and v. i. kelemen , j. phys .
b * 43 * , 045202 ( 2010 ) .
j. c. slater , _ the self - consistent field for molecules and solids _
( mcgraw - hill , new york , 1974 ) . m. ya . amusia and v. k. dolmatov , j. phys .
b * 26 * , 1425 ( 1993 ) . v. k. dolmatov , j. phys .
b * 26 * , l79 ( 1993 ) .
v. k. dolmatov , j. phys .
b * 26 * , l393 ( 1993 ) . v. k. dolmatov , j. phys .
b * 29 * , l673 ( 1996 ) .
a. a. radtsig and b. m. smirnov , _ parameters of atoms and atomic ions _ [ in russian ] ( energoatomizdat , moscow 1986 ) .
l. d. landau and e. m. lifshitz , _ quantum mechanics : nonrelativistic theory _
( butterworth - heinemann , boston 2002 ) .
e. p. wigner , phys . rev . a * 98 * , 145 ( 1955 ) . | the impacts of both exchange interaction and electron correlation , as well as their combined impact , on electron elastic scattering off a semifilled shell mn( ... @xmath0@xmath1 , @xmath2 ) atom are theoretically studied in the electron energy range of @xmath3 ev . corresponding elastic scattering phase shifts @xmath4 as well as
partial @xmath5 and total @xmath6 cross sections are found to be subject to a strong correlation impact .
the latter is shown to be drastically different for oppositely spin - polarized scattered electrons , in some cases , thereby bringing significant differences in corresponding @xmath4s , @xmath5s , and @xmath6s between said electrons .
this is proven to be an inherent features of electron scattering off a semifilled shell atom in general .
electron correlation is accounted for in the framework of the self - energy part @xmath7 of the green function of a scattered electron concept .
the latter is calculated both in the second - order perturbation theory in the coulomb interelectron interaction as well as beyond it by solving the dyson equation for @xmath7 .
the significance of the `` dyson '' correlation corrections in e@xmath8mn scattering is unraveled .
they are shown to aggravate noticeably the inherent differences between elastic scattering phase shifts and cross sections of spin - up ( @xmath9 ) and spin - down ( @xmath10 ) polarized electrons scattered off a spin - polarized mn atom , in some cases .
in particular , the existence of a narrow resonant maximum in @xmath11 near @xmath12 ev but the absence of such in @xmath13 in e@xmath8mn scattering is predicted . |
element abundances in the early universe are important observables to determine the features of the physical processes at the epoch of big bang nucleosynthesis ( bbn ) . from a discovery of expansion of universe @xcite and the development of the cosmological theory @xcite ,
it was suggested that the early universes was enough hot and dense enough to allow nuclear fusion to be operative , which is the original idea of bbn @xcite .
it was assumed that radiative neutron capture reaction generates primordial elements in hot and dense conditions .
however , neutrons and protons can be converted to each other via weak interactions .
hayashi studied this effect and proposed that bbn started from not a pure neutron gas , but a hybrid gas composed of neutrons and protons as well as radiation @xcite .
hayashi studied routes of @xmath1he production and indicated that a relic abundance of @xmath1he depends on an initial abundance of the neutron @xcite .
this suggestion and the consideration of relativistic quantum statistics refined the bbn model @xcite .
the discovery of the cosmic microwave background ( cmb ) @xcite indicated that the early universes was hot and dense , and bbn theory was confirmed .
production of d , @xmath3he , @xmath1he , @xmath2li @xcite , all stable isotopes of li , be and b @xcite , and heavier elements up to @xmath4ne @xcite have been studied . now
the bbn theory includes many important physics , and the numerical precision of the bbn calculations is rather high .
the baryon to photon ratio , @xmath0 , is one of the important parameters characterizing the bbn model .
recently , we have constrained @xmath0 from observations of cmb @xcite and tested various models of bbn using the constrained value of @xmath0 as an input .
effects of a primordial magnetic field ( pmf ) on bbn have been studied @xcite .
the pmf directly affects rates of weak reactions @xcite , and also the cosmic expansion rate through the pmf energy density . results of bbn could , therefore , be different from that for the case with no pmf @xcite .
the boost of cosmic expansion by the pmf energy density is the dominant effect of the pmf , and the change of weak reaction rates is subdominant @xcite
. an updated result of light elements up to li in the bbn model including a pmf has been derived based on a network calculation consistently taking account of important effects of pmf @xcite .
a power law ( pl ) spectrum is one of the most familiar distributions for the energy densities of various physical phenomenon on cosmological scales , including the pmf(e.g . @xcite and references therein ) .
thus many researchers have been studying the effect of the pmf with the pl spectrum on various physical processes in the early universe @xcite and have tried constraining parameters of such pmfs . in this article , providing the pl as the distribution function of the pmf ( pl - pmf ) , we report effects of the pl - pmf on primordial abundances of light elements up to li as a function of the baryon to photon ratio @xmath0 .
the energy density of the pl - pmf is dependent on the field strength , the pl index , and the maximum and minimum scales of the pmfs at the generation epochs .
therefore , we also investigate degeneracies of these pl - pmf parameters in the effects of the pmf on light element abundances in detail for the first time . in sec.ii
, we describe how to include effects of the pl - pmf on bbn in a code for the standard bbn ( sbbn ) model . in sec.iii , we release the numerical computational results of bbn with the pl - pmf , and we discuss comparisons between our results and the observational constraints on elemental abundances in sec.iv . in sec.v , we summarize our study and mention future works .
in this section , we describe our code for the bbn calculation with recent updates , and suggest how to consider the effects of the pl - pmf on bbn .
we only consider the effect of boosts of cosmic expansion by the pmf energy density because this effect is more important than other effects such as changes in the rate of weak @xcite and nuclear @xcite reactions , and those in the distribution functions of electrons and positrons @xcite , which were taken into account in ref .
we modify the kawano s bbn code @xcite , taking into consideration the pmf effects , and use the updated reaction rates with the recommendations of the jina reaclib database v1.0 @xcite as adopted in ref .
we use @xmath5 s as the neutron lifetime @xcite , which is estimated with an improved measurement @xcite .
this lifetime is somewhat shorter than the average neutron lifetime of @xmath6 s calculated by the particle data group @xcite .
it better satisfies the unitarity test of the cabibbo - kobayashi - maskawa matrix @xcite , and it slightly affects the primordial abundances ( most significantly that of @xmath1he ) @xcite .
effects of the difference between this lifetime and a longer one @xmath7 s from the old recommendation by the particle data group @xcite are seen in figs .
1 and 2 in ref . @xcite .
the uncertainties in the calculated abundances originating from an uncertainty in the neutron lifetime are much smaller than both uncertainties in observationally induced abundances and changes in calculated abundances caused by the magnetic field .
we , therefore , do not take into account the uncertainty in the neutron lifetime .
we shall describe constraints on the primordial elemental abundances that we adopt in this article .
the primordial deuterium abundance is derived by observing the lyman-@xmath8 absorption systems of the quasi - stellar objects ( qsos ) .
a recent measurement of the deuterium abundance in a damped lyman alpha system of qso sdss j1419 + 0829 was performed more precisely than those of any other qso absorption system @xcite .
we then adopt two constraints : ( 1 ) a mean value of ten qso absorption systems including j1419 + 0829 , and ( 2 ) the abundance of j1419 + 0829 from the best measurement , as follows , @xmath9 the two groups it10 @xcite and aos10 @xcite have reported different results for the primordial abundance of helium , @xmath10 , from surveys of metal - poor extragalactic hii regions .
they obtained the following constraints , i.e. , @xmath11 observations of the metal - poor halo stars ( mphss ) provide a constraint on the abundance of @xmath2li , i.e. , @xmath12 the wilkinson microwave anisotropy probe ( wmap ) project , which observes the cmb across the full sky , constrains the baryon density parameter @xcite , @xmath13 ( @xmath14 ) .
this result corresponds to the baryon to photon number ratio , @xmath15 ( @xmath14 ) . in this subsection , at first , we introduce the cosmic expansion .
next we describe an effect of the cosmic expansion on bbn .
tthe hubble parameter @xmath16 for the homogeneous and isotropic flat - universe is derived by the friedmann equation and the conservation of energy momentum tensor , @xmath17 where the dot expresses the time derivative as @xmath18 , @xmath19 is the gravitational constant , @xmath20 is the photon speed , @xmath21 and @xmath22 are the energy density and the pressure , respectively , in the universe .
the inverse of @xmath16 corresponds to the age of the universe , @xmath23 .
we assume that the main components of the universe are photon , neutrino , baryon and cold dark matter ( cdm ) , they are the ideal fluid , and the universe expands adiabatically . we describe a rate of reaction between some particles as @xmath24 .
the time scale of the reaction is then @xmath25 .
when the reaction time scale becomes larger than the age of the universe , i.e. , @xmath23 , the reaction effectively stops in the universe . whether a reaction is operative or not is determined by the following condition , i.e. , @xmath26 since the final nuclear abundances of bbn depend on when the respective reactions become ineffective , the energy densities of the universe influence the abundances of the elements produced in bbn .
we will explain how the pl - pmf energy density affects the cosmic expansion and bbn .
in the early universe including the bbn epoch , we can assume that statistically average motions of fluids are negligibly small .
in this case , from maxwell s equations and ohm s law , we derive the induction equation @xcite @xmath27 here @xmath28 is the magnetic diffusivity , which is defined by @xmath29 where @xmath30 is the electric conductivity and @xmath31 is the electrical resistance @xcite as follow , @xmath32 here @xmath33 is the mass of the electron , @xmath34 is the electron charge , @xmath35 is the thomson scattering cross section , @xmath36 is the number density of the photon , @xmath37 is the number density of the electron , and @xmath0 is the baryon to photon ratio .
equation ( [ eq_mo ] ) describes the dissipation of the magnetic field , and it shows that the magnetic field dissipates rapidly with time on the scale @xmath38 @xcite , where @xmath39 is the temperature , and @xmath40 is the age of the universe which is defined by @xmath41 s at the relativistic degrees of freedom , @xmath42 .
therefore , the magnetic field is hard to be generated on the scale @xmath43 . on the other hand ,
the magnetic field on a scale much larger than @xmath44 is hard to dissipate in the universe by time @xmath40 , and the magnetic field is `` frozen - in '' the dominant fluids @xcite . in this article
, we are interested in magnetic fields which survive until the end of bbn . in the epoch during which the electron positron annihilation has been completed ( @xmath45mev ) ,
the cutoff scale for pmf generations , @xmath44 for @xmath46 , is @xmath47 the comoving scale @xmath48 is @xmath49 and the comoving wave number @xmath50 is @xmath51 here we use @xmath52 .
thus we will report results of the effects on bbn from the pmf with a comoving wave number less than @xmath53 .
the pl spectrum is one of the most familiar forms for distributions of the energy densities including that of the pmf @xcite .
it appears in various physical phenomena operating on the cosmological scale .
many researchers have been studying the effect of the pmf with the pl spectrum on various physical processes in the early universe @xcite , and have tried to constrain parameters of such pmfs .
we then assume that the pmfs with the pl spectrum are statistically homogeneous , isotropic , and random , as assumed in previous studies , and we research their effects on bbn .
after a magnetic field is generated , the total energy density originating from that of the magnetic field is conserved , even if part of the magnetic field energy is transferred to other physical degrees of freedom ( e.g. kinetic or / and thermal energies of dominant fluids or the gravitational wave ) in the adiabatic universe .
the cosmic expansion depends on the total energy density in the universe .
therefore , when we consider the effect of the pmf energy density on the cosmic expansion , it is reasonable to define the total energy density originating from the pmf at its generational epoch .
we shall discuss cases of finite energy transfer from the pmf to other physical degrees of freedom in sec .
[ sec : discussions ] .
since the strength of the frozen - in pmf is proportional to an inverse square of the scale factor , its energy density and pressure depend on the inverse fourth power of the scale factor as follows @xcite : @xmath54 we can neglect the pressure and energy density of matter in the radiation dominated era .
therefore , the total energy density and pressure in the early universe are @xmath55 next , we shall derive the energy density of the pl - pmf .
a two - point correlation function of the strength of the pmf @xcite is given by @xmath56}(k)p^{i}_j(k)\delta(\mbi{k}-\mbi{k}')~. \label{eq : two_point_correlations_of_pmf}\end{aligned}\ ] ] here @xmath57 and @xmath58}(k ) = ak^{n_\mathrm{b}}. \label{eq_power1}\end{aligned}\ ] ] the convention for the fourier transform is @xmath59 in this article we need the energy density of the pmf in the universe , @xmath60 . from eq .
( [ eq : two_point_correlations_of_pmf ] ) , we obtain @xmath61}(k)\delta(\mbi{k}-\mbi{k } ' ) \nonumber \\ & = & 2(2\pi)^3 a k^{n_\mathrm{b}}\delta(\mbi{k}-\mbi{k } ' ) .
\label{eq : two_point_correlations_of_pmf2}\end{aligned}\ ] ] we derive the coefficient from the variance of the magnetic fields @xmath62 in real space .
first we define @xmath63 here @xmath64 is a comoving radius for a gaussian sphere , and @xmath65 is a comoving pmf strength which is scaled to the present - day value on some length scale @xmath64 . from eqs .
( [ eq : two_point_correlations_of_pmf ] ) - ( [ eq_power1 ] ) and ( [ eq_real ] ) , @xmath66 where @xmath67 is a window function here assumed to be a gaussian filter , @xmath68 .
we then obtain the coefficient @xmath62 as follows @xmath69}^{n_\mathrm{b}+3 } \gamma \left ( \frac{n_\mathrm{b}+3}{2 } \right ) } , \label{eq_a}\end{aligned}\ ] ] where @xmath70}$ ] .
therefore @xmath58}(k ) = \frac { ( 2\pi)^2 b^2_\lambda \lambda^{n_\mathrm{b}+3 } } { 2 \gamma \left ( \frac{n_\mathrm{b}+3}{2 } \right ) } k^{n_\mathrm{b } } \label{eq : two_point_correlations_of_pmf3}\end{aligned}\ ] ] from eqs ( [ eq : two_point_correlations_of_pmf ] ) - ( [ eq : two_point_correlations_of_pmf2 ] ) and ( [ eq : two_point_correlations_of_pmf3 ] ) , the energy density of the pmf is @xmath71}}_{k_\mathrm{[min ] } } \frac{dk}{k } \frac{k^3}{2\pi^2 } p_\mathrm{[pmf]}(k ) \nonumber \\ & = & \frac{2}{8\pi } \int^{k_\mathrm{[max]}}_{k_\mathrm{[min ] } } \frac{dk}{k } \frac{k^3}{2\pi^2 } \frac { ( 2\pi)^2 b^2_\lambda \lambda^{n_\mathrm{b}+3 } } { 2 \gamma \left ( \frac{n_\mathrm{b}+3}{2 } \right ) } k^{n_\mathrm{b } } \nonumber \\ & = & \frac{1}{8\pi } \frac { b^2_\lambda } { \gamma \left ( \frac{n_\mathrm{b}+5}{2 } \right ) } \left [ ( \lambda k_\mathrm{[max]})^{n_\mathrm{b}+3 } - ( \lambda k_\mathrm{[min]})^{n_\mathrm{b}+3 } \right ] , \nonumber \\ \label{eq : pl_pmf_energy_density}\end{aligned}\ ] ] where @xmath72}$ ] and @xmath73}$ ] are the maximum and minimum wave numbers , respectively , which depend on generation models .
the first purpose of this study is to investigate the effects of the pmf energy density on bbn and to constrain the pmf energy density in the bbn epoch .
the second purpose is to discuss of the degeneracy between the pl - pmf parameters and distribution models of the pmf .
to effectively proceed with these studies and discussions , we redefine the scale invariant ( si ) field strength of the pmf from eq .
( [ eq : pl_pmf_energy_density ] ) , which is constrained from observational constraints on elemental abundances , i.e. , @xmath74})^{n_\mathrm{b}+3 } - ( \lambda k_\mathrm{[min]})^{n_\mathrm{b}+3 } \right ] } { \gamma \left ( \frac{n_\mathrm{b}+5}{2 } \right ) } } , \end{aligned}\ ] ] and @xmath75 } , k_\mathrm{[min ] } ) = b^\mathrm{x}_\mathrm{si } \sqrt { \frac { \gamma \left(\frac{n_\mathrm{b}+5}{2}\right ) } { \left ( k_\mathrm{[max ] } ^{n_\mathrm{b}+3 } - k_\mathrm{[min ] } ^{n_\mathrm{b}+3 } \right ) \lambda^{n_\mathrm{b}+3 } } } , \nonumber \\ \label{eq : field_strength_for_elements}\end{aligned}\ ] ] where x is a species of light element produced in bbn and @xmath76 represents the corresponding value of the si pmf strength . since @xmath76 is proportional to @xmath77 and
does not depend on other pmf parameters , it is useful to directly understand effects of the pl - pmf amplitude on bbn
. furthermore it is relatively easy to discuss the degeneracy of the pmf parameters if we use this formalization .
in this section , we show calculated results of the bbn with the pl - pmf using our code . we also show how the strength of the pl - pmf is constrained from limits on abundances of light elements up to li .
figure [ fig1 ] shows abundances of @xmath1he ( @xmath78 : mass fraction ) , d / h , @xmath3he / h , @xmath2li / h and @xmath79li / h as a function of @xmath0 for cases of bbn with ( solid lines ) and without ( dashed lines ) the effects of the pl - pmf . the parameters of the pl - pmf are set to be @xmath80 g , @xmath81 , @xmath82}=10^{-13}$ ] @xmath83 , and @xmath84}=0 $ ] @xmath83 .
this parameter set corresponds to a si pmf strength of @xmath85 g .
the vertical painted band shows the limit on the baryon to photon ratio derived from wmap 7yr data @xcite .
the horizontal painted bands and lines with downward arrows indicate observational constraints and upper limits , respectively , on elemental abundances .
effects of the pmf on abundances of respective nuclides have been analyzed in detail @xcite .
the effects are described below to explain the abundances predicted as a function of @xmath0 in fig .
[ fig1 ] .
when the pl - pmf exists , abundances of the neutron are larger than those of the bbn without the pl - pmf , i.e. , the sbbn .
the reason is as follows : considering the energy density of the pl - pmf , the cosmic expansion rate @xmath16 is larger than that of sbbn
. therefore , the weak reaction freeze - out occurs earlier @xcite [ cf . eqs .
( [ gamma_h1 ] ) and ( [ gamma_h2 ] ) ] , and abundances of the neutron at the freeze - out increase .
furthermore , the faster expansion leads to a shorter time interval between the freeze - out and @xmath1he production , so abundances of the neutron , which is spontaneously decaying after the freeze - out , are higher at the epoch of @xmath1he production .
since @xmath1he nuclei are formed from almost all neutrons , the energy density of the pl - pmf increases the abundance of @xmath1he [ fig .
[ fig1](a ) ] . deuterium ( @xmath86h , d ) is produced and destroyed by @xmath87 the high neutron abundance in the pl - pmf case leads to an increase in d abundance because of an enhanced production rate through the reaction ( [ pn_dg ] ) [ fig . [ fig1](b ) ] .
@xmath3h and @xmath3he are produced via the reactions ( [ dd_tp ] ) and ( [ dd_3hen ] ) , and destroyed by @xmath88 the enhanced abundances of the neutron and d result in higher @xmath3h and slightly higher @xmath3he abundances than those of sbbn .
@xmath3h nuclei eventually @xmath89 decay to @xmath3he .
the final primordial abundance of @xmath3he is then the sum of those of @xmath3h and @xmath3he in the bbn epoch .
primordial @xmath3he is mainly produced as @xmath3he ( in sbbn more than 99.5 % of the final abundances of @xmath3he result from those of @xmath3he for @xmath90@xmath91 ) . the slight increase in @xmath3he at the bbn epoch
enhances the final @xmath3he abundance slightly [ fig .
[ fig1](b ) ] .
@xmath2li is mainly produced and destroyed by @xmath92 a combination of a significant increase in @xmath3h abundance and a small fractional increase in @xmath1he abundance , both by the pl - pmf , results in a significant increase in the @xmath2li production rate [ eq .
( [ 4het_7ligamma ] ) ] , while a small fractional decrease in h abundance ( its mass fraction is @xmath93 ) results in a slight decrease in the @xmath2li destruction rate [ eq .
( [ 7lip_4he4he ] ) ] . as a result , @xmath2li abundance is larger than that in sbbn .
@xmath2be is mainly produced and destroyed by @xmath94 a significant increase in neutron abundance by the pl - pmf results in an increase in the @xmath2be destruction rate [ eq .
( [ 7ben_7lip ] ) ] , while a combination of a slight increase in @xmath3he abundance and a small fractional increase in @xmath1he abundance results in a slight increase in the @xmath2be production rate [ eq .
( [ 4he3he_7begamma ] ) ] . as a result ,
@xmath2be abundance is smaller than that in sbbn .
@xmath2be nuclei eventually capture electrons and are converted to @xmath2li nuclei at the epoch of cosmological recombination .
the primordial @xmath2li abundance is , therefore , contributed from both @xmath2li and @xmath2be produced at bbn .
if the baryon to photon ratio @xmath0 is lower than some critical value @xmath95 , abundance of @xmath2li is larger than that of @xmath2be . on the other hand ,
if @xmath0 is larger than @xmath95 , abundance of @xmath2be is larger than that of @xmath2li .
therefore , the equality , i.e. , @xmath2li / h=@xmath2be / h , is satisfied at @xmath96 . since the pmf increases @xmath2li abundance , while decreases @xmath2be abundance , the final @xmath2li abundances are higher in @xmath97 and lower in @xmath98 than those in sbbn . it should be noted that the @xmath95 value and the minimum value of the final @xmath2li abundance , which is located at a valley of the curve for @xmath2li / h , are moved from those of sbbn by the pmf effects : the values of sbbn are @xmath99 and @xmath2li / h@xmath100 at @xmath101 , while those of bbn with the pl - pmf are @xmath102 and @xmath2li / h@xmath103 at @xmath104 [ fig .
[ fig1](c ) ] .
@xmath79li is produced and destroyed by @xmath105 the relatively large enhancement of d abundance and a small fractional increase in @xmath1he result in an efficient production of @xmath79li [ fig .
[ fig1](d ) ] .
the @xmath1he abundance increases as a function of @xmath0 , and it increases as a function of the pmf amplitude . the limits on @xmath1he abundance , therefore , prefer an @xmath0 value smaller than in the sbbn when the effect of the pmf is considered .
the d abundance is smaller for larger @xmath0 values , while it is larger for larger amplitudes of the pmf .
the limits on d abundance then prefer an @xmath0 value larger than in the sbbn .
the @xmath2li abundance is larger for larger @xmath0 values , while it is smaller for larger amplitudes of the pmf .
the limit on @xmath2li abundance , therefore , prefers an @xmath0 value larger than in the sbbn when the effect of the pmf is considered ( see sec .
[ subsec : discusses2 ] also ) . to research not only qualitative results but also the quantitative relationship between @xmath76 and @xmath0 ,
we show how the pl - pmf strength @xmath76 can be constrained via observational constraints on the abundance of light elements [ eqs .
( [ limited_d_mean ] ) ( [ limited_d ] ) , ( [ limited_yp_it ] ) , ( [ limited_yp_aos ] ) and ( [ limited_7li ] ) ] and the @xmath0 value . figure [ fig2 ] shows calculated constraints of @xmath76s as a function of @xmath0 .
the vertical painted band is the limit on the baryon to photon ratio from the wmap 7yr data @xcite .
we use abundances of only @xmath1he , d , and @xmath2li to constrain the @xmath76 value since limits on @xmath76 from abundances of @xmath3he and @xmath79li are less severe than those of @xmath1he , d , and @xmath2li [ figs .
[ fig1](b ) and ( d ) ] .
four kinds of lines in fig .
[ fig2](a ) correspond to upper limits , i.e. , @xmath76 , from @xmath10 @xmath106 0.2777 ( aos10 ; black - thin - dotted - dashed line ) , @xmath10 @xmath106 0.2667 ( it10 ; green - thick - dotted - dashed line ) , d / h @xmath106 2.88 @xmath107 ( mean ; blue - thin line ) , d / h @xmath106 2.64 @xmath107 ( best ; red - thick line ) , and @xmath2li / h @xmath106 2.35@xmath108 ( purple - dotted line ) , respectively . on the other hand , requiring a consistency between @xmath2li abundances of theory and observations leads to a lower limit on the pl - pmf as a function of @xmath0 [ the right purple - dotted line in fig .
[ fig2](a ) ] because the abundance of @xmath2li decreases when the energy density of the pl - pmf increases ( sec.[subsec : bbnwithpmf ] ) .
then , the allowed region from the @xmath2li limit on the @xmath109 vs. @xmath0 plane is between the two purple - dotted curves in fig .
[ fig2](a ) . the limits on pmf strengths , i.e , @xmath76 , at the @xmath0 value of a 2 @xmath30 lower limit by the wmap 7yr data @xcite , i.e , @xmath0=5.916 @xmath108 , are @xmath110
since the abundance of d / h is already more than d / h = @xmath111 ( best ) when no pmf is assumed , we can not put a constraint ( @xmath112 ) using d / h @xmath113 at @xmath114 , when we adopt the @xmath0 values of the best - fit and 2 @xmath30 upper limit by the wmap 7yr data @xcite , constraints from @xmath2li abundance are weaker , or @xmath109 values are larger .
we then show four constraints ( @xmath76 ) , except for that of @xmath2li as follows : @xmath115 and @xmath116 this result is very similar to the latest result @xcite of the bbn calculation which consistently includes effects of the pmf through changes in the distribution functions of electrons and positrons as well as the cosmic expansion rate .
this is because the effect of the pmf on bbn through changes in distribution functions is smaller than that through the cosmic expansion rate @xcite . adopting our results [ ( [ lower_b_yp_it])-([upper_b_d2 ] ) and fig . [ fig2 ] ] and the wmap-@xmath0 value @xcite , we show combined constraints on the pl - pmf [ fig .
[ fig2](b ) ] .
if we adopt limits on abundances of @xmath10(it10 ) and d / h ( mean ) , we obtain a combined upper limit : @xmath117 the maximum value corresponds to the limit at @xmath118 ( see the intersection of the thin solid curve and the thick dot - dashed curve in fig . [ fig2 ] ) .
if we adopt @xmath10 ( it10 ) and d / h ( best ) , the following upper limit is derived : @xmath119 the maximum value corresponds to the limit at @xmath120 ( see the intersection of the thick solid curve and the vertical painted band in fig .
[ fig2 ] ) .
in the above sections , we explained the effect of pl - pmf on bbn and derived constraints on pmf energy densities . in this section , considering these results , we will discuss degeneracy between the pl - pmf parameters , ( @xmath121},~ k_\mathrm{[min]}$ ] ) in effects on bbn , how such constrained pl - pmf parameters are applied for constraint on pmf generation models , and cases of an energy transferring from pmf to others physical degrees of freedom .
the pl - pmf strength @xmath122 depends on @xmath72}$ ] , @xmath73}$ ] and @xmath123 [ eq . ( [ eq : field_strength_for_elements ] ) ] .
in particular , if @xmath73}/ k_\mathrm{[max]}$ ] is close to 1 or / and @xmath124 is close to -3 @xcite , the influence of @xmath73}$ ] on @xmath122 can not be ignored and we should treat the @xmath73}$ ] value carefully in constraining pmf parameters and generation models ( fig . [ fig3 ] ) , although @xmath73}$ ] does not affect the energy density of the pl - pmf otherwise .
if @xmath73}/k_\mathrm{[max]}$ ] is close to 1 , such a pmf is generated on narrow scale ranges . in this case , the distribution of the pmf may be better described by the gaussian , delta or log - normal @xcite functions than the pl .
when @xmath123 is close to -3 , the pmf is expected to be generated in the epoch of inflation @xcite . in this case , from eq . ( [ eq : field_strength_for_elements ] ) and fig . [ fig3 ]
( b ) , the term of @xmath73}^{n_\mathrm{b}+3}$ ] can not be neglected even if @xmath125 is relatively small .
thus we need to be careful in constraining the pmf of inflationary origin .
we have discussed the specific cases of the pl - pmf .
next , we discuss the opposite cases in which ranges of wave number for the generated pl - pmf spectrum are wide enough so that @xmath73}$ ] can be neglected ( @xmath73}/k_\mathrm{[max]}\ll1 $ ] ) .
figure [ fig4 ] shows constraints on the field strength @xmath65 as a function of @xmath123 [ cf .
( [ eq : field_strength_for_elements ] ) and ( [ upper_b_yp_it1 ] ) ] for fixed values of @xmath126 mpc and @xmath73 } = 0 $ ] .
it is clear that the limits on this plane of @xmath65 vs. @xmath123 from light element abundances , i.e. , their negative slopes , show a negative correlation of the parameters [ eq . ( [ eq : field_strength_for_elements ] ) ] in a parameter region for the fixed si strength of the pmf .
when generated pl - pmfs have spectra with smaller @xmath72}$ ] , constraints on the ( @xmath65 , @xmath123 ) plane are more stringent . in general , considering the cosmic expansion after the inflation , the smaller the generation scale of the magnetic field , the earlier the magnetic field is generated .
this feature is useful in constraining models for pmf generation in the post - inflation epoch .
since cosmological observations have constrained strengths of the pl - pmf @xmath65 at @xmath127 mpc that are less than orders of 1 ng @xcite , we shall discuss the degeneracy between @xmath72}$ ] and @xmath123 in the pmf effect on bbn for 0.1 ng @xmath128 10 ng .
figure [ fig5 ] shows constraints on the maximum wave number @xmath72}$ ] as a function of @xmath123 [ cf .
( [ eq : field_strength_for_elements ] ) and ( [ upper_b_yp_it1 ] ) ] for fixed values of @xmath126 mpc and @xmath73 } = 0 $ ] . results for various @xmath65 values of 10 ng ( dashed line ) , 1 ng ( solid line ) and 0.1 ng ( dot - dashed line ) are shown .
it is obvious that the limits on the plane of @xmath72}$ ] and @xmath123 show a negative correlation of the parameters [ eq . ( [ eq : field_strength_for_elements ] ) ] in a parameter region for the fixed si strength of the pmf .
the cosmological observations ( e.g. the cmb and the large scale structures : see refs .
@xcite for details ) have constrained the pmf with strength of the order of 1 ng at 1 mpc . as mentioned in sec .
[ subsec : models_ed_plpmf ] , we are interested in the magnetic field on scales much larger than the cutoff scale for the pmf which is generated before the end of the bbn epoch , @xmath129 . in the case where @xmath123 is less than -2.6 and @xmath130 ng , the @xmath72}$ ] value is allowed to be larger than @xmath131 cm @xmath132 ( fig .
[ fig5 ] ) . on the other hand ,
if @xmath123 is larger than -2.6 , the allowed value of @xmath72}$ ] becomes smaller .
however the upper limits on @xmath72}$ ] for very large @xmath123 are smaller than the cosmic expansion rate , i.e. , @xmath16 ( the lower limits on @xmath72}^{-1}$ ] are larger than the horizon scale ) before bbn , and such a pmf can only be generated before the end of inflation .
therefore , using this feature , the pl index of the pmf generated in the post - inflation epoch can be strongly constrained . from fig .
[ fig2 ] , it is confirmed that this bbn model including the effect of the pmf energy density can not be a solution to the discrepancy between @xmath2li abundances produced in sbbn and those deduced from observations of mphss .
we consider a case where part of the pmf energy ( @xmath133 ) is dissipated and transferred to the radiation energies .
the final value of the photon energy density measured at an epoch of scale factor @xmath134 is then @xmath135 , where @xmath136 is a primary photon energy density which does not include that transferred from the pmf by dissipation .
since @xmath137 increases as a function of time from the time of the pmf generation , the following equation always holds : @xmath138 .
this means that the baryon to photon ratio at the bbn epoch is expected to be more than the present value , and the @xmath0 value in the bbn epoch constrained from the wmap 7yr data would be larger than the case of no energy transfer . here , we conservatively consider the most general case in which the history of generation and dissipation of the pmf in the universe is completely unrestricted .
the amplitude of the pmf in the bbn epoch and the largeness of the pmf energy dissipation during a time from bbn to the cosmological recombination should then be regarded as two independent parameters .
the effect of the pmf on the cosmic expansion rate increases the abundances of @xmath1he , d and @xmath3he , and decreases that of @xmath2li for @xmath0 values around the wmap 7yr estimation for the sbbn ( @xmath139 ) .
the effect of dissipation , on the other hand , increases abundances of @xmath1he and @xmath2li and decreases those of d and @xmath3he ( figs .
[ fig1 ] and [ fig2 ] ) .
effects of the pmf and dissipation on the abundances of d , @xmath3he and @xmath2li work in opposite directions , while they both increase @xmath1he abundance .
one can , therefore , understand that an upper limit on the strength of the pmf from @xmath1he , i.e. , @xmath140 , can be derived by assuming that a pmf dissipation does not occur . in this regard ,
the case of fig .
[ fig1 ] corresponds to the maximum amplitude of the pmf allowed by the constraint from @xmath1he abundance for @xmath141 .
the minimum @xmath2li abundance in the bbn model with a free parameter of @xmath142 and @xmath141 is given by the cross region of the solid line for @xmath2li / h and the vertical band ( fig .
[ fig1 ] ) , which is higher than the abundances in mphss .
we then conclude that the abundances of @xmath1he and @xmath2li can not be simultaneously consistent with observations [ eqs .
( [ limited_yp_aos ] ) and ( [ limited_7li ] ) ] , even in the case where both the energy density and the dissipation of the pmf are taken into account .
we showed quantitative and qualitative effects of a pmf on each abundance of a light element using a bbn model taking account of the pmf with two parameters of the si strengths of pmfs and the baryon to photon ratio @xmath0 ( figs .
[ fig1 ] and [ fig2 ] ) .
we assumed a pmf characterized by the power spectrum of a pl in wave number .
parameters of this pl - pmf are a field amplitude @xmath65 , a pl index @xmath123 , and maximum and minimum scales at a generation epoch @xmath72}$ ] and @xmath73}$ ] , respectively .
we then showed a relation between pl - pmf parameters and the si strength of the pmf which is the parameter constrained from observations of light element abundances ( sec .
[ subsec : models_ed_plpmf ] ) .
we then calculated bbn taking into account recent updates on reaction rates and the effect of the pmf , and showed results of the abundances in the example case of ( @xmath143 } , k_{\rm [ min]})=(2.2~\mu$]g , @xmath144 , @xmath131 @xmath83 , @xmath145 @xmath83 ) as a function of @xmath0 .
it was observed that the critical @xmath0 value , i.e. , @xmath95 , at which @xmath2li and @xmath2be are produced in the same abundance at the bbn epoch , moves by pmf effects .
the value @xmath99 for sbbn is enhanced up to @xmath102 for the example case .
in addition , the minimum of the final @xmath2li abundance , located at the valley of the @xmath2li / h vs. @xmath0 curve moves from @xmath2li / h@xmath100 at @xmath101 for the sbbn to @xmath2li / h@xmath103 at @xmath104 for bbn with the pl - pmf ( sec .
[ subsec : bbnwithpmf ] ) .
we investigated constraints on parameters of the pl - pmf from the abundances of light elements up to li produced in bbn .
we showed respective and combined constraints on the parameters . as a result
, we obtained upper limits on the si strength of the pmf , i.e. , @xmath146 g from abundances of @xmath1he and d for the 2 @xmath30 region of @xmath0 values by the wmap 7yr data ( sec .
[ subsec : discussions2 ] ) . in sec.iv , we discussed the degeneracy of the pl - pmf parameters in effects on bbn and showed some possibility of constraining models of pmf generations by combining constraints on the si field strength from light element abundances to the relation between pl - pmf parameters and the si field strength ( sec . [ subsec : discussions1 ] ) .
in addition , we consider a general case in which the existence and energy dissipation of the pmf are allowed . based on our result of the bbn calculation ( figs .
[ fig1 ] and [ fig2 ] ) , it was found that an upper limit on the strength of the pmf can be derived from a constraint on @xmath1he abundance .
a lower limit on @xmath2li abundance is also derived , and it is significantly higher than those observed in metal - poor stars .
we then conclude that it is impossible to solve the @xmath2li problem by the pmf energy density , even if we consider that part of the pmf energy is dissipated and transferred to the radiation energies ( sec .
[ subsec : discusses2 ] ) . in the near future
, cosmological observations of the cmb and large scale structures will make it possible to constrain the pl - pmf parameters more tightly .
if we combine such results and our constraint on the parameters through the study of bbn , we will not only limit the pl - pmf parameters but also the generation models of the pmf more precisely .
abundances of @xmath10(mass fraction of @xmath1he ) , d / h , @xmath3he / h , @xmath2li / h and @xmath79li / h as a function of @xmath0 in bbn models with and without the effect of the pl - pmf for the fixed value of @xmath126 mpc .
the vertical painted band is the limit on the baryon to photon ratio from wmap 7yr data @xcite .
the white dashed line in the limited range of @xmath0 indicates the best value .
the horizontal painted bands and lines with downward arrows indicate observational constraints and upper limits , respectively , on elemental abundances .
the constraints on @xmath3he and @xmath79li abundances are from ref .
@xcite and refs .
@xcite , respectively.,scaledwidth=50.0% ] constraints on the field strength @xmath76 derived from the consistency of the abundances of light elements , d , @xmath1he and @xmath2li .
we use the following upper limits on the abundances : @xmath10 = 0.2667 ( thick dot - dashed solid line ) and 0.2777 ( thin dote - dashed solid line ) , d / h = 2.88@xmath107 ( thin solid line ) and 2.64@xmath107 ( thick solid line ) , and @xmath2li / h = 2.35@xmath108 ( dotted line ) [ panel ( a ) ] .
the vertical painted band is the limit on the baryon to photon ratio from wmap 7yr data @xcite .
the white dashed line indicates the best value .
the right panel shows an enlarged view of allowed regions constrained from observations of cmb and elemental abundances .
painted regions in panel ( b ) are as indicated in the legend box at the top.,scaledwidth=100.0% ] ratio of the magnetic field
@xmath147}\ne 0)/ b_\lambda(k_\mathrm{[min]}=0)$ ] as a function of @xmath123 [ cf .
eqs.([eq : field_strength_for_elements ] ) and ( [ upper_b_yp_it1 ] ) ] for fixed ratios of wave number @xmath73}/k_\mathrm{[max]}$ ] as shown in the legend box ( a ) .
the right panel shows @xmath147}\ne 0)/ b_\lambda(k_\mathrm{[min]}=0)$ ] as a function of @xmath73}/k_\mathrm{[max]}$ ] for fixed values of @xmath123 ( b).,scaledwidth=100.0% ] constraints on the field strength @xmath148 as a function of @xmath123 [ cf .
eqs.([eq : field_strength_for_elements ] ) and ( [ upper_b_yp_it1 ] ) ] for fixed values of @xmath126mpc and @xmath73 } = 0 $ ] .
the thin ( red ) , thick ( green ) , and dashed - thick ( blue ) lines show upper limits for cases of @xmath72 } = 10^{-13}$ ] , @xmath149 , and @xmath150 , respectively.,scaledwidth=100.0% ] constraints on the maximum wave number @xmath72}$ ] as a function of @xmath123 [ cf .
eqs.([eq : field_strength_for_elements ] ) and ( [ upper_b_yp_it1 ] ) ] for fixed values of @xmath126mpc and @xmath73 } = 0 $ ]
. results for various @xmath65 values of 10 ng ( dashed line ) , 1 ng ( solid line ) and 0.1 ng ( dot - dashed line ) are shown .
, scaledwidth=100.0% ] | big bang nucleosynthesis ( bbn ) is affected by the energy density of a primordial magnetic field ( pmf ) . for an easy derivation of constraints on models for pmf generations
, we assume a pmf with a power law ( pl ) distribution in wave number defined with a field strength , a pl index , and maximum and minimum scales at a generation epoch .
we then show a relation between pl - pmf parameters and the scale invariant ( si ) strength of pmf for the first time .
we perform a bbn calculation including pmf effects , and show abundances as a function of baryon to photon ratio @xmath0 .
the si strength of the pmf is constrained from observational constraints on abundances of @xmath1he and d. the minimum abundance of @xmath2li / h as a function of @xmath0 slightly moves to a higher @xmath2li / h value at a larger @xmath0 value when a pmf exists during bbn .
we then discuss degeneracies between the pl - pmf parameters in the pmf effect .
in addition , we assume a general case in which both the existence and the dissipation of pmf are possible . it is then found that an upper limit on the si strength of the pmf can be derived from a constraint on @xmath1he abundance , and that a lower limit on the allowed @xmath2li abundance is significantly higher than those observed in metal - poor stars . |
the spectroscopy of light mesons plays an important role in understanding the strong interactions at low energies . among possibilities to study the spectrum of light mesons , analysis of the @xmath1 interaction
is particularly useful and , therefore , it has always been an object of continuous theoretical and experimental investigation @xcite . here , we present results of the coupled - channel analysis of data on processes @xmath2 in the channels with @xmath3 and @xmath4 and on the @xmath1 scattering in the channel with @xmath5 .
the scalar sector is problematic up to now especially as to an assignment of the discovered mesonic states to quark - model configurations in spite of a big amount of work devoted to these problems ( see , _ e.g. _ , ref .
@xcite and references therein ) .
an exceptional interest to this sector is supported by the fact that there , possibly indeed , we deal with a glueball @xmath6 ( see , _
e.g. _ , ref .
@xcite ) .
investigation of vector mesons is up - to - date subject due to their role in forming the electromagnetic structure of particles and because our knowledge about these mesons is still too incomplete ( _ e.g. _ , in the particle data group tables @xcite ( pdg ) the mass of @xmath7 is ranging from 1250 to 1582 mev ) . in the tensor sector , among the thirteen discussed resonances , the nine states ( @xmath8 , @xmath9 , @xmath10 , @xmath11 , @xmath12 , @xmath13 , @xmath14 , @xmath15 , @xmath16 ) must be confirmed in various experiments and analyses . for example , in the analysis of @xmath17 , five resonances @xmath18 , @xmath13 , @xmath14 , @xmath19 and @xmath20 have been obtained , one of which , @xmath13 , is a candidate for the glueball @xcite . in our analysis , we have used both _
a model - independent method _
@xcite , based on the first principles ( analyticity and unitarity ) directly applied to analysis of experimental data , and _ the multichannel breit
wigner forms_. the former approach permits us to introduce no theoretical prejudice to extracted parameters of resonances , however , it is limited with the possibility to use only three coupled channels . therefore , in more general cases , one has to use , _
e.g. _ , the breit
wigner approach .
considering the obtained disposition of resonance poles on the riemann surface , obtained coupling constants with channels , and resonance masses we draw particular conclusions about nature of the investigated states .
in both methods of analysis , we parametrized the @xmath21-matrix elements @xmath22 where @xmath23 denote channels , using the le couteur - newton relations @xcite .
this relations express the @xmath21-matrix elements of all coupled processes in terms of the jost matrix determinant @xmath24 that is a real analytic function with the only square - root branch - points at the channel momenta @xmath25 . in the model - independent approach
, the @xmath21-matrix is determined on the 4- and 8-sheeted riemann surfaces for the 2- and 3-channel cases , respectively .
the matrix elements @xmath22 have the right - hand cuts along the real axis of the @xmath26 complex plane ( @xmath26 is the invariant total energy squared ) , starting at the coupled - channels thresholds @xmath27 ( @xmath28 ) , and the left - hand cuts related to the crossed channels .
the riemann - surface sheets are numbered according to the signs of analytic continuations of the channel momenta @xmath29 , as shown in table [ ksigns ] .
.signs of channel momenta on the eight sheets of the rieman surface in the 3-channel case . [ cols="^,^,^,^,^,^,^,^,^",options="header " , ]
in analysis of the processes @xmath30 , we considered explicitly also the channel @xmath31 . here
it is impossible to use the uniformizing - variable method .
therefore , using the le couteur - newton relations , we generate the resonance poles by some 4-channel breit - wigner forms .
the @xmath32-function is taken as @xmath33 , where the resonance part is @xmath34\ ] ] with @xmath35 and @xmath36 the partial width .
the blatt
weisskopf barrier factor for a tensor particle is @xmath37 with radii of 0.943 fm for all resonances in all channels , except for @xmath38 and @xmath39 for which they are : for @xmath38 , 1.498 , 0.708 , and 0.606 fm in the channels @xmath1 , @xmath40 , and @xmath41 , respectively ; for @xmath39 , 0.296 fm in the channel @xmath40 . the background part has the form @xmath42\ ] ] with @xmath43 @xmath44 gev@xmath45 is a combined threshold of the channels @xmath46 , @xmath47 , and @xmath48 .
the data for the @xmath1 scattering are taken from an energy - independent analysis by hyams _
et al . _
the data for @xmath49 are taken from works @xcite .
we obtained a satisfactory description with ten resonances @xmath38 , @xmath8 , @xmath50 , @xmath51 , @xmath52 , @xmath11 , @xmath39 , @xmath13 , @xmath19 , and @xmath53 ( the total @xmath54 ) and with eleven states adding one more resonance @xmath14 which is needed in the combined analysis of processes @xmath17 @xcite . in our analysis ,
the description with eleven resonances is practically the same as that with ten resonances : the total @xmath55 .
the obtained resonance parameters are shown in table [ tab : tensor_param ] for the cases of ten and eleven states .
ccccccc state & @xmath56 & @xmath57 & @xmath58 & @xmath59 & @xmath60 & @xmath61 + + & [email protected] & [email protected] & [email protected] & [email protected] & [email protected] & @xmath63212 + @xmath8 & [email protected] & [email protected] & 562.3@xmath62142 & [email protected] & 8.2@xmath6265 & @xmath64230 + @xmath50 & [email protected] & [email protected] & 253.8@xmath6278 & [email protected] & 41.6@xmath62160 & @xmath6449 + @xmath9 & [email protected] & [email protected] & [email protected] & [email protected] & 127@xmath62199 & @xmath64170 + @xmath52 & [email protected] & 78.8@xmath6243 & [email protected] & [email protected] & [email protected] & @xmath64182 + @xmath11 & [email protected] & [email protected] & [email protected] & [email protected] & 90.3@xmath6290 & @xmath64177 + @xmath39 & [email protected] & [email protected] & [email protected] & [email protected] & 65.4@xmath6294 & @xmath64119 + @xmath13 & [email protected] & [email protected] & [email protected] & 58.8@xmath6224 & 450.4@xmath62221 & @xmath64299 + @xmath19 & [email protected] & [email protected] & 551@xmath62149 & 375@xmath62114 & 166.8@xmath62104 & @xmath64222 + @xmath53 & [email protected] & [email protected] & [email protected] & [email protected] & 460.8@xmath62209 & @xmath64170 + + & [email protected] & [email protected] & [email protected] & [email protected] & [email protected] & @xmath63210.5 + @xmath8 & [email protected] & [email protected] & 562.3@xmath62144 & [email protected] & 8.2@xmath6263 & @xmath64230 + @xmath50 & [email protected] & [email protected] & 253.9@xmath6279 & [email protected] & 51.6@xmath62155 & @xmath6449.5 + @xmath9 & [email protected] & [email protected] & [email protected] & [email protected] & 127@xmath62190 & @xmath64170 + @xmath52 & [email protected] & 78.8@xmath6243 & [email protected] & 460.3@xmath62545 . & 108.6@xmath6276 .
& @xmath64182.4 + @xmath11 & [email protected] & [email protected] & 259@xmath6232 . &
[email protected] & [email protected] & @xmath64177.6 + @xmath39 & [email protected] & [email protected] & [email protected] & [email protected] & [email protected] & @xmath64118.6 + @xmath13 & [email protected] & [email protected] & [email protected] & [email protected] & 574.8@xmath62211 & @xmath64193 + @xmath14 & [email protected] & [email protected] & [email protected] & [email protected] & 128@xmath62190 & @xmath64107 + @xmath19 & [email protected] & [email protected] & [email protected] & 381@xmath62116 & 168.8@xmath62103 & @xmath64222 + @xmath53 & [email protected] & [email protected] & [email protected] & [email protected] & 462.8@xmath62211 & @xmath64168 the background parameters for ten resonances are : @xmath65 , @xmath66 , @xmath67 , @xmath68 , @xmath69 , @xmath70 , @xmath71 , @xmath72 , @xmath73 ; and for eleven resonances are : @xmath74 , @xmath75 , @xmath76 , @xmath77 , @xmath78 , @xmath79 , @xmath80 , @xmath81 , @xmath82 . in figures
[ y12 ] and [ y34 ] we show results of fitting to the data . -scattering @xmath83-wave @xmath21-matrix element.,title="fig:",scaledwidth=84.0% ] -scattering @xmath83-wave @xmath21-matrix element.,title="fig:",scaledwidth=84.0% ] ( upper figure ) and @xmath84 ( lower figure ) @xmath83-wave
@xmath21-matrix elements.,title="fig:",scaledwidth=83.0% ] ( upper figure ) and @xmath84 ( lower figure ) @xmath83-wave
@xmath21-matrix elements.,title="fig:",scaledwidth=83.0% ]
in the combined model - independent analysis of data on the @xmath2 processes in the channel with @xmath3 , an additional confirmation of the @xmath85-meson with mass 835 mev is obtained ( the pole position on sheet ii is @xmath86 mev ) .
this value of mass corresponds most near to the one ( @xmath87 mev ) of ref .
@xcite and rather accords with prediction ( @xmath88 ) on the basis of mended symmetry by s. weinberg @xcite .
note that our values of @xmath89 and @xmath90 for the @xmath91-pole position are larger than those obtained in the dispersive analysis of data on only the @xmath1 scattering , see ref .
@xcite and reference therein .
indication for @xmath92 to be the @xmath41 bound state is obtained . from the point of view of the quark structure , this is the 4-quark state .
maybe , this is consistent somehow with arguments in favour of the 4-quark nature of @xmath92 @xcite .
the @xmath93 and @xmath94 have the dominant @xmath95 component .
conclusion about the @xmath93 agrees quite well with the one drawn by the crystal barrel collaboration @xcite where the @xmath93 is identified as @xmath41 resonance in the @xmath96 final state of the @xmath97 annihilation at rest .
conclusion about the @xmath98 is quite consistent with the experimental facts that this state is observed in @xmath99 @xcite and not observed in @xmath100 @xcite . as to the @xmath6
, we suppose that it is practically the eighth component of octet mixed with the glueball being dominant in this state .
its biggest width among the enclosing states tells also in behalf of its glueball nature @xcite .
we propose the following assignment of scalar mesons below 1.9 gev to lower nonets , excluding the @xmath92 as the @xmath41 bound state . :
the isovector @xmath101 , the isodoublet @xmath102 , and @xmath91 and @xmath103 as mixtures of the eighth component of octet and the su(3 ) singlet .
then the gell - mann okubo ( gm - o ) formula @xmath104 gives @xmath105 mev ( @xmath106 mev ) . in the relation for masses of nonet @xmath107
the left - hand side is about 25 % bigger than the right - hand one . , the obtained value of mass for the @xmath113 is smaller in the model - independent approach , @xmath114 mev , and a little bit bigger in the breit
wigner one , @xmath115 mev , than the averaged value cited in the pdg tables @xcite , @xmath116 mev .
however , it also occurs in analysis of some reactions ( see pdg tables ) .
the obtained value of the total width in the first case ( @xmath117 mev ) is in a good agreement with the averaged pdg one ( @xmath118 mev ) and it is a little bit bigger in the second case ( @xmath119 mev ) than the averaged pdg value , however , this is encountered also in other analyses ( see pdg tables ) .
note that predicted widths of the @xmath113 decays to the @xmath120-modes are significantly larger than , _
e.g. _ , the ones evaluated in the chiral model of some mesons based on the hidden local symmetry added with the anomalous terms @xcite .
the first @xmath121-like meson has the mass 1257.8@xmath12211 mev in the model - independent analysis and [email protected] mev in the breit
wigner one .
these values differ significantly from the mass ( 1459@xmath12211 mev ) of the first @xmath121-like meson cited in the pdg tables .
the @xmath123 was discussed actively some time ago @xcite and later the evidence for its existence was obtained in @xcite .
if the @xmath123 is interpreted as the first radial excitation of the @xmath5 @xmath124 state , then it lies down well on the corresponding linear trajectory with an universal slope on the @xmath125 plane ( n is the radial quantum number of the @xmath124 state)@xcite , whereas the @xmath7 turns out to be considerably higher than this trajectory .
the @xmath123 and the isodoublet @xmath126 are well located to the octet of the first radial excitations .
the mass of the latter should be by about 150 mev larger than the mass of the former .
then the gm - o formula @xmath127 gives @xmath128 mev , that is fairly good compatible with the mass of the first @xmath129-like meson @xmath130 , for which one obtains the values in range 1350 - 1460 mev ( see pdg tables ) .
existence of the @xmath7 ( along with @xmath123 ) does not contradict to the data . in the @xmath124 picture
, it might be the first @xmath131 state with , possibly , the isodoublet @xmath132 in the corresponding octet . from the gm - o formula ,
we should obtain the value 1750 mev for the mass of the eighth component of this octet .
this corresponds to one of the observations of the second @xmath129-like meson with masses from 1606 to 1840 mev that is cited in the pdg tables under the @xmath133 . as to the @xmath134 , in this energy region there are practically no data on the @xmath0-wave of @xmath1 scattering .
the model - independent analysis testifies in favour of existence of this state , whereas the breit wigner analysis gives the same description with and without the @xmath134 .
the suggested picture for the first two @xmath121-like mesons is consistent with predictions of the quark model @xcite . in ref .
@xcite the discussed mass spectrum for radially excited @xmath121 and @xmath135 mesons was obtained using rather simple mass operator . if the existence of the @xmath123 is confirmed , some quark potential models , _
e.g. _ , in ref .
@xcite , will require substantial revisions , because the first @xmath121-like meson is usually predicted about 200 mev higher than this state .
to the point , the first @xmath135-like meson is obtained in the indicated quark model at 1580 mev , whereas the corresponding very well established resonance has the mass of only 1410 mev . , we carried out two analysis without and with the @xmath14 .
we do not obtain @xmath10 , @xmath12 and @xmath15 , however , we see @xmath136 and @xmath52 which are related to the statistically - valued experimental points .
usually one assigns the states @xmath38 and @xmath50 to the ground tensor nonet .
to the second nonet , one could assign @xmath137 and @xmath138 though for now the isodoublet member is not discovered .
if @xmath139 is the isovector of this octet and if @xmath137 is almost its eighth component , then , from the gm - o formula , we expect this isodoublet mass at about 1633 mev .
then the relation for masses of nonet would be fulfilled with a 3% accuracy .
_ @xcite observed the strange isodoublet with yet indefinite remaining quantum numbers and with mass @xmath140 mev in the mode @xmath141 .
this state might be the tensor isodoublet of the second nonet .
the states @xmath142 and @xmath143 together with the isodoublet @xmath144 could be put into the third nonet . then in the relation for masses of nonet @xmath145 the left - hand side is only 5.3 % bigger than the right - hand one .
if one consider @xmath142 as the eighth component of octet , the gm - o formula @xmath146 gives @xmath147 mev .
this value coincides with the one for @xmath148-meson obtained in works @xcite .
this state is interpreted as a second radial excitation of the @xmath149-state on the basis of consideration of the @xmath148 trajectory on the @xmath125 plane @xcite . as to @xmath13 ,
the presence of the @xmath14 in the analysis with eleven resonances helps to interpret @xmath13 as the glueball . in the case of ten resonances ,
the ratio of the @xmath1 and @xmath41 widths is in the limits obtained in ref .
@xcite for the tensor glueball on the basis of the 1/n - expansion rules .
however , the @xmath40 width is too large for the glueball . at practically the same description of processes with the consideration of eleven resonances as in the case of ten ,
their parameters have varied a little , except for the ones for @xmath13 and @xmath53 .
mass of the latter has decreased by about 40 mev . as to @xmath13 ,
its @xmath40 width has changed significantly .
now all the obtained ratios of the partial widths are in the limits corresponding to the glueball .
9 c. amsler _
_ , phys . lett . *
b667 * , 1 ( 2008 ) .
anisovich , int . j. mod
a * 21 * , 3615 ( 2006 ) . c. amsler and f.e .
close , phys .
d * 53 * , 295 ( 1996 ) .
et al . _ , int . j. mod
a * 20 * , 6327 ( 2005 ) .
d. krupa , v.a .
meshcheryakov , and yu.s .
surovtsev , nuovo cimento a * 109 * , 281 ( 1996 ) .
le couteur , proc .
r. london , ser .
a * 256 * , 115 ( 1960 ) ; r.g .
newton , j. math .
* 2 * , 188 ( 1961 ) ; m. kato , ann .
phys . * 31 * , 130 ( 1965 ) .
d. morgan and m.r .
pennington , phys .
d * 48 * , 1185 ( 1993 ) .
b. hyams _ et al .
b * 64 * , 134 ( 1973 ) ; _ ibid . _ * 100 * , 205 ( 1975 ) .
a. zylbersztejn _
b * 38 * , 457 ( 1972 ) ; p. sonderegger , p. bonamy , in _ proc .
conference on elementary particles , lund , 1969 _ , 372 ; j.r .
et al . _ , phys .
b * 36 * , 134 ( 1971 ) ; j.p . baton _ et al .
b * 33 * , 525 ( 1970 ) ; _ ibid . _ * 33 * , 528 ( 1970 ) ; p. baillon _
et al . _ ,
b * 38 * , 555 ( 1972 ) ; l. rosselet _
et al . _ ,
d * 15 * , 574 ( 1977 ) ; a.a .
_ , pisma zh .
. fiz . * 25 * , 68 ( 1977 ) ; a.a .
belkov _ et al .
_ , pisma zh .
fiz . * 29 * , 652 ( 1979 ) .
g. grayer _
b * 75 * , 189 ( 1974 ) .
r. kaminski , l. lesniak , and k. rybicki , z. phys .
c * 74 * , 79 ( 1997 ) .
w. wetzel _
et al . _ ,
b * 115 * , 208 ( 1976 ) ; v.a .
et al . _ ,
d * 19 * , 1317 ( 1979 ) ; p. estabrooks , phys .
d * 19 * , 2678 ( 1979 ) ; d. cohen _ et al .
d * 22 * , 2595 ( 1980 ) ; g. costa _
et al . _ ,
b * 175 * , 402 ( 1980 ) ; a. etkin _
et al . _ ,
d * 25 * , 1786 ( 1982 ) .
f. binon _ et al .
_ , nuovo cimento a * 78 * , 313 ( 1983 )
. f. binon _ et al .
_ , nuovo cimento a * 80 * , 363 ( 1984 ) .
protopopescu _ et al .
d * 7 * , 1279 ( 1973 ) ; p. estabrooks and a.d .
martin , nucl .
b * 79 * , 301 ( 1974 ) .
surovtsev and p. bydovsk , nucl .
phys . a * 807 * , 145 ( 2008 ) .
j. bohacik and h. khnelt , phys .
d * 21 * , 1342 ( 1980 ) .
achasov and a.a .
kozhevnikov , phys .
d * 71 * , 034015 ( 2005 ) . v.
bernard , a.a .
osipov , and u.g .
meissner , phys .
b * 285 * , 119 ( 1992 ) .
osipov , a.e .
radzhabov , and m.k .
volkov , arxiv : hep - ph/0603130 .
i. caprini , g. colangelo , and h. leutwyler , int . j. mod
. phys . a * 21 * , 954 ( 2006 ) .
r. kamiski , l. leniak , and b. loiseau , phys .
b*551 * , 241 ( 2003 ) .
pelez and f.j .
yndurin , phys .
d * 71 * , 074016 ( 2005 ) .
s.j . lindenbaum and r.s .
longacre , phys .
b * 274 * , 492 ( 1992 ) ; r.s .
b * 177 * , 223 ( 1986 ) .
n.a . tornqvist and m. roos , phys .
lett . * 76 * , 1575 ( 1996 ) .
s. weinberg , phys .
lett . * 65 * , 1177 ( 1990 ) .
r. garca - martn , r. kamiski , and j.r .
pelaez , arxiv:0810.1134[hep - ph ] .
achasov , nucl .
a * 675 * , 279c ( 2000 ) ; m.n .
b * 438 * , 441 ( 1998 ) ; _ ibid .
_ * 440 * , 442 ( 1998 ) . c. amsler _ et al . _ ,
b * 355 * , 425 ( 1995 ) .
s. braccini , frascati phys .
series * xv * , 53 ( 1999 ) .
r. barate _
b * 472 * , 189 ( 2000 ) .
v.v . anisovich _
et al . _ ,
suppl . ) * 56 * , 270 ( 1997 ) .
budnev _ et al .
lett . b**70 * * , 365 ( 1977 ) ; s.b . gerasimov and a.b .
govorkov , z. phys .
c * 13 * , 43 ( 1982 ) ; _ ibid . _ * 29 * , 61 ( 1985 ) .
d. aston _ et al .
b ( proc . suppl . )
* 21 * , 105 ( 1991 ) ; t.s .
belozerova and v.k .
henner , phys .
elem . part .
* 29 * , 148 ( 1998 ) ; yu.s .
surovtsev and p. bydovsk , arxiv : hep - ph/0701274 , frascati phys .
series , vol .
xlvi , 1535 ( 2007 ) ; i. yamauchi , t. komada , frascati phys .
series , vol .
xlvi , 445 ( 2007 ) .
anisovich , v.v .
anisovich , and a.v .
sarantsev , phys .
d * 62 * , 051502 ( 2000 ) .
e. van beveren , g. rupp , t.a .
rijken , and c. dullemond , phys .
d * 27 * , 1527 ( 1983 ) . s.b . gerasimov and a.b .
govorkov , z. phys .
c * 13 * , 43 ( 1982 ) ; _ ibid . _ * 29 * , 61 ( 1985 ) .
s. godfrey and n. isgur , phys .
d * 32 * , 189 ( 1985 ) .
v.m . karnaukhov _
et al . _ , yad . fiz . *
63 * , 652 ( 2000 ) .
a.v . anisovich _
et al . _ ,
b * 452 * , 173 ( 1999 ) ; _ ibid .
_ * 452 * , 187 ( 1999 ) ; _ ibid . _
* 517 * , 261 ( 2001 ) . | in the analysis a status and parameters of the scalar , vector , and tensor mesonic resonances are obtained and compared with other results . possible classification of the resonance states in terms of the su(3 ) multiplets is discussed .
* outline : * * motivation * method of analysis * analysis of the isoscalar - scalar sector * analysis of the isovector @xmath0-wave of @xmath1 scattering * analysis of the isoscalar - tensor sector * spectroscopic implications from the analysis |
in a number of extremal problems in probability and statistics ( see e.g. @xcite ) one needs to evaluate the positive - part moments @xmath0 of random variables ( r.v.s ) @xmath2 , where @xmath3 , the positive part of @xmath4 , and @xmath5 , for any @xmath6 and @xmath1 .
in particular , an effective procedure was needed in @xcite to compute @xmath0 for @xmath7 and r.v.s of the form @xmath8 , where @xmath9 are positive real numbers , @xmath10 , @xmath11 and @xmath12 are independent r.v.s , @xmath11 has the normal distribution with parameters @xmath13 and @xmath14 , and @xmath12 has the poisson distribution with parameter @xmath15 . for purely normal r.v.s @xmath2 ( without the poisson component ) such
a computation is easy .
however , the naive approach , by the formula @xmath16 did not work well , especially when @xmath17 is small . on the other hand , the series of the form @xmath18 ( which is the laplace - fourier transform of the poisson distribution ) is easily computable .
therefore , a natural idea was to perform a harmonic analysis of the function @xmath19 .
this can be easily done by finding its laplace - fourier transform .
then , once the function @xmath20 is decomposed into harmonics ( i.e. , exponential functions @xmath21 for @xmath22 ) , one almost immediately obtains a general expression for the positive - part moments of a r.v .
@xmath2 in terms of the laplace - fourier transform of the distribution of @xmath2 , as provided indeed by theorem [ lem : ga ] in this paper .
such expressions turn out to be computationally effective , as well as the ones in terms of the characteristic function ( c.f . )
@xmath23 that are obtained as corollaries .
theorem [ cor : improp ] of this paper provides a necessary and sufficient condition for such a representation .
a subsequent literature search has revealed only one paper , brown @xcite , that contains formulas for @xmath0 in terms of the c.f . of @xmath2 .
however , the constant - sign argument used in @xcite does not actually seem to work for @xmath24 .
the results in @xcite are based on the method developed by von bahr @xcite to express the absolute moments @xmath25 in terms of the c.f . of @xmath2 .
other papers containing such expressions for absolute moments include @xcite ; see also @xcite and @xcite .
as was explained , our approach differs from the previous ones .
we begin with expressions of positive - part moments in terms in the fourier - laplace transform @xmath26 with @xmath27 , and then use the cauchy integral theorem to express @xmath0 in terms of the c.f . of @xmath2 .
it should be clear that expressions for the absolute moments will follow trivially from expressions for the positive - part moments .
let @xmath2 be any r.v .
let @xmath28 and @xmath29 be any real numbers such that @xmath30 and @xmath31 for all @xmath32 $ ] .
let @xmath33 be any positive real number , and then let @xmath34 respectively the integer part of @xmath33 and the smallest integer that is no less than @xmath35 , so that @xmath36 , @xmath37 , and @xmath38 .
for all complex @xmath39 and all @xmath40 , let @xmath41 with the convention @xmath42 for any @xmath43 s , so that @xmath44 , @xmath45 , and @xmath46 .
for any complex number @xmath47 , where @xmath48 and @xmath49 are real numbers and @xmath50 stands for the imaginary unit , let @xmath51 and @xmath52 , the real and imaginary parts of @xmath39 . in this paper , we shall present a number of identities involving certain integrals . for
the sake of brevity , let us assume the following convention ( unless specified otherwise ) : when saying that such an identity takes place under certain conditions , we shall actually mean to say that under those conditions the corresponding integral exists in the lebesgue sense ( but may perhaps be infinite ) _ and _ the identity takes place . [
lem : ga ] for any @xmath53 $ ] and any @xmath54 such that @xmath55 , @xmath56 [ rem : details ] of course , the statement of theorem [ lem : ga ] is devoid of content in the case when @xmath57 . as for the condition @xmath55
, here we use the convention @xmath58 ( any other real number in place of @xmath59 would do here as well ) , to interpret @xmath60 when @xmath61 and @xmath62 .
so , the condition @xmath55 will trivially hold if @xmath62 . on the other hand , if @xmath63 ( and hence @xmath64 ) , then the expression @xmath65 under the integrals on the right - hand sides of and would lose meaning without the condition @xmath55 .
[ cor : neg ] if @xmath66 and @xmath67 , then for any @xmath68 and any @xmath54@xmath69 [ cor:0 ] if @xmath67 then @xmath70 in particular , one has the following : for any @xmath71 such that @xmath72 , @xmath73 for any @xmath71 such that @xmath74 , @xmath75 [ cor : xy ] if @xmath76 is another r.v
. then @xmath77 provided that @xmath78 and @xmath79 for all @xmath80 .
moreover , one has the following : for any @xmath71 @xmath81 if @xmath82 and @xmath83 for all @xmath84 ; for any @xmath71 @xmath85 if @xmath86 and @xmath87 for all @xmath84 . for any r.v .
@xmath2 as in corollary [ cor:0 ] , it is always possible ( and easy ) to construct a r.v .
@xmath76 as in corollary [ cor : xy ] such that the characteristic function @xmath88 , and hence its real and imaginary parts @xmath89 and @xmath90 , are easily computable
. in particular , one can always take @xmath76 to be a r.v . with finitely many values
; more specifically , @xmath91 values would suffice for and @xmath92 values for or .
an obvious advantage of the formulas given in corollary [ cor : xy ] ( over those in corollary [ cor:0 ] ) is that the corresponding integrals will converge ( for @xmath64 ) faster near @xmath93 , and just as fast near @xmath94 .
also , corollary [ cor : xy ] will be just one step closer than corollary [ cor:0 ] to possible applications to the convergence of the positive - part moments in the central limit theorem ; see example [ ex : clt ] in section [ appls ] for further details .
[ rem : left ] if @xmath95 for an even natural @xmath96 , then it is clear that identity ( or , equivalently , ) can not hold without the condition @xmath67 , since one needs the moment @xmath97 on the right - hand side of to exist .
similarly , if @xmath95 for an odd natural @xmath98 , then identity ( or , equivalently , ) can not hold without the condition @xmath67 . however , if @xmath1 is not an integer , the term @xmath99 on the right - hand side of disappears .
one may then wonder as to what , if any , moment condition on the _ left _ tail of the distribution of @xmath2 is needed in order for identity to hold having in mind that the finiteness of the positive part moment @xmath0 depends only on the _ right _ tail of the distribution of @xmath2 . perhaps surprisingly , it turns out that a left - tail condition is necessary for , even if the integral in is allowed to be understood as an improper one ( at @xmath94 ) ; moreover , the moment condition @xmath67 in corollary [ cor:0 ] can be only slightly relaxed , even when @xmath33 is not an integer .
actually , one can obtain a necessary and sufficient condition for such a representation of the positive - part moment in terms of the characteristic function , as described in the following theorem .
[ cor : improp ] take any @xmath100 and any r.v .
@xmath2 with @xmath95 .
then the following two conditions are equivalent to each other : 1 .
[ improp ] @xmath101and @xmath102 2 .
[ improp equiv ] @xmath103 here and in what follows , @xmath104 for all @xmath6 ; also let @xmath105 .
note that the part @xmath101 of condition ( i ) of theorem [ cor : improp ] will trivially hold if @xmath106 ( and hence @xmath107 ) . on the other hand , if @xmath64 ( and hence @xmath108 ) , then the expression @xmath109 in condition ( i ) of theorem [ cor : improp ] would lose meaning if @xmath110 ; cf .
remark [ rem : details ] .
note also that , if condition ( ii ) or , equivalently , ( i ) of theorem [ cor : improp ] holds , then @xmath111 for all @xmath112 .
indeed , condition ( ii ) of theorem [ cor : improp ] implies @xmath113 on the other hand , it is easy to give examples where condition ( ii ) of theorem [ cor : improp ] holds , while @xmath114 , and so , corollary [ cor:0 ] is not applicable ; for instance , take any r.v .
@xmath2 such that @xmath115 for all large enough @xmath116 . in view of corollary [ cor:0 ] and
theorem [ cor : improp ] , for the identity to hold for a given @xmath100 ( with the integral understood in the lebesgue sense ) and a r.v .
@xmath2 with @xmath95 , it is sufficient that @xmath117 and it is necessary that @xmath111 for all @xmath112 .
yet , no `` simple '' necessary _ and _ sufficient condition for to hold in the lebesgue sense for a given @xmath100 appears to be known ; it is apparently an open question whether such a condition exists at all .
however , one can see that condition ( ii ) of theorem [ cor : improp ] is not sufficient ( although , by theorem [ cor : improp ] , it is necessary ) for .
indeed , for any @xmath100 , there exists a r.v .
@xmath2 with @xmath95 such that the two equivalent conditions ( i ) and ( ii ) of theorem [ cor : improp ] hold , while the integral @xmath118 does not exist in the lebesgue sense .
for instance , one can consider [ ex : improp ] the idea is to take a r.v .
@xmath2 with a lacunary distribution and then construct a matching lacunary set @xmath119 such that the integrand @xmath120 is of constant sign and large enough in absolute value for @xmath121 .
take indeed any @xmath100 and let @xmath2 be a discrete r.v . such that , for some @xmath122 , one has @xmath123 where @xmath124 is the constant chosen so that @xmath125
. then it is easy to check that condition ( ii ) ( and hence condition ( i ) ) of theorem [ cor : improp ] holds .
moreover , @xmath126 a.s . , and so , @xmath95
further , one can show that there exist some @xmath127 , @xmath128 , and @xmath129 such that @xmath130 for all @xmath131 $ ] , whence @xmath132 . for details , see ( * ? ? ?
* proposition 2.6 and its proof ) .
theorem [ cor : improp ] and example [ ex : improp ] are similar in spirit to a number of known if - and - only - if conditions and counterexamples , respectively ; those results in the literature mainly concern such simpler relations as the ones of the tails / moments of the distribution with the derivatives of the characteristic function at @xmath94 ( but also with other linear functionals of the c.f . )
see e.g. @xcite .
ibragimov @xcite provides necessary and sufficient conditions both in terms of ( truncated ) moments / tails and the c.f .
for rates @xmath133 in the central limit theorem to hold ; cf .
example [ ex : clt ] in the next section .
here let us give examples of known or potential applications of identities presented in section [ results ] .
[ ex : hoeff ] let @xmath134 be independent r.v.s such that @xmath135 for some real @xmath136 and @xmath137 , for all @xmath50 .
let @xmath138 and assume that @xmath139 for some real @xmath140 .
also , let @xmath141 , so that @xmath142 . in the paper @xcite , mentioned in the introduction , an optimal in a certain sense upper bound on the tail @xmath143 was obtained , of the form @xmath144 where @xmath6 ; @xmath145 ; @xmath146 and @xmath147 are any independent r.v.s whose distributions are , respectively , the centered gaussian one with variance @xmath148 and the centered poisson one with variance @xmath149 ; @xmath150 is the only real root of the equation @xmath151 ; and @xmath152 .
thus , to compute the bound @xmath153 , one needs a fast calculation of the positive - part moments of the r.v . @xmath154 . while the direct approaches mentioned in the introduction do not work here ,
the formulas of theorem [ lem : ga ] and corollary [ cor:0 ] prove very effective , since the fourier - laplace transform of the r.v .
@xmath155 is given by a simple expression : @xmath156 it takes under @xmath157 sec in mathematica on a standard core 2 duo laptop to produce an entire graph of the bound @xmath153 for a range of values of @xmath4 using either or .
[ ex : finance ] as pointed out by a referee , positive - part moments naturally arise in mathematical finance .
for example , @xmath158 is the value of a call option with the strike price @xmath159 when the stock price is @xmath160 .
exact bounds on @xmath161 under various conditions were recently given in @xcite .
[ ex : clt ] just as it was done in @xcite for absolute moments , the identities given in corollary [ cor : xy ] could be used to prove the convergence of the positive - part moments in the central limit theorem , or even to obtain bounds on the rate of such convergence .
however , we shall not be pursuing this matter here , since the main motivation for the present paper was the need for an effective computation of the bound @xmath153 , as described in example [ ex : hoeff ] .
for any two expressions @xmath162 and @xmath163 , let us write : @xmath164[0pt]{$\frown$}}}}{<}}$}}}}b$ ] or , equivalently , @xmath165[0pt]{$\frown$}}}}{>}}$}}}}a$ ] if @xmath166 for some positive constant @xmath167 depending only on @xmath33 ; @xmath168 if @xmath169 , @xmath164[0pt]{$\frown$}}}}{<}}$}}}}b$ ] , and @xmath165[0pt]{$\frown$}}}}{<}}$}}}}a$ ] .
let us also write @xmath170 ( to be read `` @xmath162 is much less than @xmath163 '' ) or , equivalently , @xmath171 ( `` @xmath163 is much greater than @xmath162 '' ) if @xmath172 and @xmath173 .
accordingly , if @xmath174 is an assertion and @xmath175 , then `` @xmath174 holds if @xmath170 '' will mean `` for all @xmath100 there exists some constant @xmath176 such that @xmath174 holds whenever @xmath177 '' , that is , @xmath174 will hold `` eventually '' . similarly , `` if @xmath170 and @xmath174 holds , then @xmath178 '' will mean `` for all @xmath100 and all @xmath179 there exists @xmath180 such that one has the implication ` if @xmath177 and @xmath174 holds , then @xmath181 ' '' . note that immediately follows from , since the integrand in is even in @xmath184 .
note also that @xmath185 for all @xmath186 and @xmath187 , where @xmath188 .
so , it is enough to prove with @xmath189 .
the key observation here is the following : for any @xmath6 and any complex number @xmath47 with @xmath190 , @xmath191 this is obvious for real @xmath192 , and then one can use analytic continuation . thus , for each @xmath193 the fourier transform @xmath194 of the integrable function @xmath195 is the function @xmath196 , which is integrable as well .
so , by the inverse fourier transform , one obtains ( with @xmath189 ) with the real constant @xmath4 in place of the r.v .
( equivalently , one can use here the laplace inversion . )
now follows by the fubini theorem , since @xmath197 for all @xmath198 and @xmath184 . here , just as in the proof of theorem [ lem : ga ] , without loss of generality ( w.l.o.g . )
consider first the case when @xmath199 .
, @xmath200 . by the convexity of @xmath201 in @xmath48 ,
one has @xmath202 for all @xmath32 $ ] and all @xmath203 , so that @xmath204 as @xmath205 .
hence , by the cauchy integral theorem and the continuity of the integral @xmath206 in @xmath32 $ ] , the difference between the value of @xmath207 for @xmath53 $ ] and that of @xmath207 for @xmath68 is @xmath208 , where @xmath209 is the circle in @xmath210 of radius @xmath211 centered at @xmath94 and traced out counterclockwise , provided that @xmath212 .
so , ( in the case when @xmath199 and with @xmath189 ) follows from by taking @xmath213 , since @xmath214 over @xmath215 ; recall that here it is assumed that @xmath66 .
assume now that @xmath57 ( and , again , @xmath189 ) .
then , by what has just been proved , one has for any @xmath68 with @xmath216 in place of @xmath2 , where @xmath217 is any real number . by the dominated convergence theorem and in view of the condition @xmath67 , one has @xmath218 and @xmath219 as @xmath220 . to complete the proof of corollary [ cor : neg ]
, it remains to use again the dominated convergence theorem , in view of the inequalities @xmath221 and @xmath222 for all @xmath223 , @xmath68 , and @xmath184 .
let us first prove identity in the case when the r.v .
@xmath2 takes on only one value , say @xmath6 .
for any @xmath224 , consider the integration contour @xmath225 , where @xmath226 .
then , by ( with @xmath189 ) and the cauchy integral theorem , @xmath227 because @xmath228 for all @xmath186 .
next , @xmath229 as @xmath230 , and so , @xmath231 as @xmath213 . this and imply @xmath232 since the latter integrand is even in @xmath184 . the case @xmath66 is treated quite similarly .
namely , if @xmath96 for some @xmath71 , then @xmath236 for all @xmath6 and @xmath237 , and so , follows by the fubini theorem from , the estimate @xmath238[0pt]{$\frown$}}}}{<}}$}}}}|u|^{2m-1}\wedge|u|^{2m+1}$ ] for all @xmath239 , and the condition @xmath72 .
if @xmath98 for some @xmath71 , then similarly follows using the estimate @xmath240[0pt]{$\frown$}}}}{<}}$}}}}|u|^{2m-2}\wedge|u|^{2m}$ ] for all @xmath239 and the condition @xmath74 .
_ for @xmath246 , let @xmath247 this definition is correct , since the integral exists in the lebesgue sense , in view of estimate . note that @xmath248 this is a special case of , with @xmath249 almost surely ( a.s . ) . in turn , implies @xmath250 _ _ step 2.__at this step , let us prove that if @xmath251 and @xmath252 , then @xmath253 take indeed any @xmath251 .
then @xmath107 , and and can be rewritten as @xmath254 for all @xmath255 .
observe next that @xmath256 for all @xmath255 , whence the only local minima of @xmath257 in @xmath258 are at the points @xmath259 .
therefore , and because of and the obvious equality @xmath260 , to prove it suffices to show that @xmath261 for all @xmath262 . using and ( twice ) applying the fubini theorem , for all @xmath262 and @xmath263 such that @xmath264 one has @xmath265 letting now @xmath266 and using again the equality @xmath260 , by the dominated convergence and a comparison of the integrands one obtains @xmath267 which completes the proof of .
_ step 3 . _ still assuming that @xmath251 , it follows from that @xmath268[0pt]{$\frown$}}}}{>}}$}}}}j^{-p}$ ] over all @xmath262 .
next , for every @xmath262 , the only local minima of @xmath257 on the interval @xmath269 $ ] are the endpoints ; so , @xmath270[0pt]{$\frown$}}}}{>}}$}}}}j^{-p}{\mathrel{\text{\raisebox{2pt}{${\underset{{\text{\raisebox{3pt}[0pt][0pt]{$\frown$}}}}{>}}$}}}}v^{-p}$ ] over all @xmath271 $ ] and all @xmath262 .
that is , @xmath272[0pt]{$\frown$}}}}{>}}$}}}}v^{-p}$ ] over all @xmath273 . on the other hand , by , @xmath274[0pt]{$\frown$}}}}{<}}$}}}}v^{-p}$ ] over all @xmath252 .
thus , @xmath275 over all @xmath273 .
_ _ step 4.__here , let us show , by induction on @xmath280 , that for each @xmath100 @xmath281 over all @xmath255 and @xmath116 , where @xmath282 for all @xmath255 and @xmath283
. for @xmath107 ( that is , for @xmath251 ) , is equivalent to .
assume now that @xmath108 and holds for @xmath284 and @xmath35 instead of @xmath280 and @xmath33 , respectively .
then , by and fubini s theorem , for all @xmath255 and @xmath116 @xmath285 where @xmath286 for @xmath237 , so that @xmath287 over all @xmath49 in a right neighborhood of @xmath94 , @xmath288 over all @xmath49 in a left neighborhood of @xmath93 , and @xmath289 on @xmath258
. therefore , @xmath290 over all @xmath237 .
this and yield .
_ _ step 5.__in view of and since @xmath291 for all real @xmath49 and @xmath4 , one has @xmath292 for all @xmath255 by fubini s theorem , which is applicable because of and . by with @xmath293 in place of @xmath2
, the integral on the right - hand side of converges to @xmath294 as @xmath295 .
vice versa , suppose now that condition ( ii ) holds .
then for any @xmath300 and all real @xmath301 @xmath302 as @xmath303 .
also , for all @xmath116 @xmath304 as @xmath303 .
so , condition ( ii ) implies .
this completes step 6 and thus the entire proof of theorem [ cor : improp ] . , _ binomial upper bounds on generalized moments and tail probabilities of ( super)martingales with differences bounded from above _ , in ims lecture notes - monograph series , high dimensional probability , institute of mathematical statistics , * 51 * , ( 2006 ) .
doi : 10.1214/074921706000000743 . | integral expressions for positive - part moments @xmath0 ( @xmath1 ) of random variables @xmath2 are presented , in terms of the fourier - laplace or fourier transforms of the distribution of @xmath2 . a necessary and sufficient condition for the validity of such an expression
is given .
this study was motivated by extremal problems in probability and statistics , where one needs to evaluate such positive - part moments .
address : department of mathematical sciences + michigan technological university + houghton , michigan 49931 , usa + e - mail : ` [email protected] ` + telephone : 906 - 487 - 2108 running head : positive - part moments via the fourier - laplace transform |
hundreds of brown dwarfs ( bds ) have been discovered in the last few years , both within star - forming regions @xcite and among field stars @xcite . still
very little is known about their formation processes .
first indications that bds may form as ordinary stars , i.e. via accretion from a circumstellar disk , come from observations of h@xmath0 @xcite and near - infrared excess emission @xcite .
however , in the case of bds only excess emission at longer wavelengths provides the indisputable evidence for the presence of circumstellar material @xcite . thermal emission of warm dust around bds has been observed by a number of groups @xcite .
very recently , our search for submm / mm emission from young bds provided the first detection of cold dust around two of them @xcite .
for the first time , this allowed to estimate the mass of the circumstellar material and to relate it to the mass and luminosity of the central object .
these results open up new interesting questions : _ is the circumstellar material around bds confined in a disk ?
are these disks similar to t tauri disks or different ones ? what is the effect of the ultra - low luminosity on the disk properties ? _ tentative modeling of disks has been performed by @xcite based on two isocam mid - infrared ( mir ) measurements with the general assumption of flared disk geometry .
the first ground - based observations of one of their objects showed that the flared disk model can not be applied blindly : at least in one case a flat configuration can better explain the observations @xcite .
the same flat geometry has been later used by @xcite to interpret the isocam measurements of other bds .
a thorough investigation of the disk properties requires measurements of different disk zones , i.e. in different wavelength domains . in this letter
we compile the first complete spectral energy distribution ( sed ) of a disk around a bd .
the goal of this paper is to test the basic models and determine which model parameters can be constrained from the combination of short- and long - wavelengths observations .
our study focuses on the object cfht - bd - tau 4 ( 2massi j0439474 + 260140 ) which was found and confirmed as a young bd by @xcite .
its very strong h@xmath0 emission @xcite coupled with x - ray activity @xcite and further spectral signatures @xcite prove that this bd is intensively accreting matter . in the following ,
we first present the sed of cfht - bd - tau 4 , we prove that the circumstellar material can not be in a spherical shell and we argue towards a disk configuration .
we model the sed applying different disk geometries .
finally , we discuss similarities and differences with accretion disks around t tauri stars .
in this section we compile the observations available on cfht - bd - tau 4 .
optical and near - infrared data , together with an estimate of the line - of - sight extinction ( a@xmath1=3 mag ) , have been published by @xcite .
an additional l - band flux is given in @xcite .
we retrieved archival isocam images at 6.7 and 14.3 and we measured fluxes of 28.4 and 38.5 mjy by using an aperture of 24 .
we accept a photometric accuracy of 20% as given in the isocam calibration accuracies document .
our group observed cfht - bd - tau 4 in the mm and submm continuum at the iram-30 m and the jcmt telescopes .
the data reduction and analysis is presented in @xcite .
we provide the summary of the above fluxes and errors in table [ flux ] .
the main input parameters for the modeling are derived directly from the observations .
cfht - bd - tau 4 is expected to be a member of the taurus star - forming region located at a distance of 140 pc .
we assume an effective temperature of @xmath2 k for an [email protected] spectral type @xcite and a mass of 70 m@xmath4 . in agreement with @xcite , we calculate a total luminosity of 0.1 l@xmath5 which is brighter than the luminosity of the youngest ( 1 myr ) isochrone from @xcite .
the wavelength dependence of the dust opacity ( @xmath6 ) and the mass of the circumstellar dust can be calculated from the expected optically thin submm and mm fluxes . in the optically thin regime
the flux depends on the wavelengths as f@xmath7 , where @xmath8 is the wavelength dependence of @xmath6 .
a value of @xmath9 2 has been measured for interstellar dust particles @xcite , while @xmath9 1 is generally used for circumstellar disks where grain growth is expected to take place @xcite . in our case
the slope of the sed is @xmath10 , with the error computed from the uncertainty in the flux measurements .
this implies that @xmath8 is @xmath11 . due to the large error on the measured @xmath8
, we can not fix the dust opacity . to compare our results with disks around t tauri stars
, we assume the same dust opacity as suggested by @xcite which amounts to 2 @xmath12/g at 1.3 mm . for a dust temperature between 10 and 20 k and a gas - to - dust ratio of 100
, we calculate a total circumstellar mass of 1.50.3 m@xmath4 .
we note that adopting the 10 times lower opacity of interstellar grains would lead to 10 times larger circumstellar masses .
a simple spherical shell model is first investigated by using the radiative transfer ( rt ) code of @xcite .
the mass of the circumstellar material is set to 4 m@xmath4 in order to reproduce the submm and mm fluxes .
different power - law density distributions , inner / outer radii and dust properties were tested ( see table [ models ] ) .
the 1d simulations always predict a visual extinction at least an order of magnitude higher than measured by @xcite and @xcite .
thus , the simple shell configuration has to be excluded and a 2d distribution has to be considered . the most plausible 2d circumstellar dust distribution is believed to be in the form of a disk @xcite . in the following , we test various versions of the most widely accepted disk configurations , i.e. the flat @xcite and the flared disk @xcite .
the opening angle of the flat disk is constant , while that of the flared disk increases with radius .
an optically thin superheated layer is expected for flared configurations .
an addition to some of the flared disks is the self - consistent treatment of the inner rim that , due to its higher temperature , puffs - up with respect to the flaring disk behind it @xcite .
similar changes in the flared disk geometry have been proposed by @xcite as an effect of stellar accretion and viscous dissipation .
for all the models , we adopt a dust opacity which follows the silicate mass absorption coefficient by @xcite for short wavelengths and the opacity law by @xcite for long wavelengths .
the transition wavelength depends on the grain size : for small grains of 0.1 radius the two opacity laws meet at 100 , while for 10 grain they meet at 200 .
we consider two flat disk models ( m4 and m5 in table [ models ] ) : one with a single disk temperature distribution @xcite and one with an additional hot surface layer .
the main difference between the two approximations is at 10 , where the two - layer model predicts an emission feature , which is absent in the optically thick one - layer flat disk . for both configurations , plausible physical parameters
provide a good fit . because of the lack of measurements at the silicate feature
, we can not discriminate between the two cases .
for the flared disk configurations @xcite , a too high mir flux is predicted by the models when the disk inner radius is determined by the dust sublimation temperature . to reduce the mir emission three possibilities
are considered : high disk inclination , a large inner gap and the shadow of a puffed - up inner rim .
considering the low line - of - sight extinction measured towards cfht - bd - tau 4 , almost edge - on disks ( inclinations larger than 60 ) are not plausible . even when a large inner gap of about 0.07 au is introduced in the classical flared disk model ( m7 ) , the mir measurements at 6.7 and 14.3 can not be simultaneously fitted . on the contrary
, the flared disk with a puffed - up inner rim and large inner gap reproduces well the sed of cfht - bd - tau 4 ( m9 in table [ models ] and solid line in fig .
[ sed ] ) . to demonstrate the effect of grain growth , we also tested an opacity law for silicate grains with 10 radius @xcite .
such grains lack the emission feature at @xmath13 .
the one - layer flat disk model is not affected by the different dust opacities .
for the two - layer flat configuration , the silicate feature at 10 vanishes , but the sed can still be well fitted with parameters similar to those provided in table [ models ] . because larger grains have lower opacities in the mir regime , the flared disk model with the puffed - up rim can fit the observations with a slightly smaller inner gap ( 0.04 au ) .
even with the addition of larger grains , the classical flared disk model does not explain the observed sed .
the combination of short and long wavelengths observations allow us , for the first time , to investigate the properties of the circumstellar dust around a bd .
the sed shows an overall shape similar to that of t tauri stars . adopting the opacity by @xcite
, we calculate a circumstellar mass of 0.31.5 m@xmath4 , which is a few percent of the bd s mass . a similar mass ratio is found for t tauri stars and their disks .
thus , the disk is unlikely to be truncated by any previous ejection event @xcite .
a discussion is presented in @xcite . since spherically distributed dust would result in too high extinction , we argue towards the presence of a circumstellar disk .
our disk models suggest that _ either the disk is flat , or it is flared with a puffed - up inner rim and a large inner gap_. a flat geometry was previously suggested by @xcite to explain the sed of a disk around a bd candidate .
the present results support a flat disk as well , thus suggesting that this might be general for disks around very low - luminosity objects .
a flat dust distribution could be a hint towards large grain sizes since large grains are expected to settle in the disk midplane ( see @xcite and @xcite for a discussion of the physical conditions under which flared disks can exist ) .
future observations at 10 could further strengthen this picture if no silicate feature is found .
should the disk be flared with a puffed - up inner rim , an inner gap 5 times larger than the dust sublimation radius is required to account for the observed sed .
the presence of magnetic fields is not sufficient to explain inner gaps as large as those found in the modeling ( about 9 r@xmath5 ) , see for example @xcite .
the material close to the bd could be cleared out by a massive planet , as suggested in the case of tw hya by @xcite
. however , the most realistic explanation is the presence of a close companion with a spectral type similar to the primary .
this would also explain the over - luminosity of this bd @xcite .
unfortunately , the estimated separation of the possible companion would be @xmath14 mas , unresolvable with any current imaging technique . to fully characterize bd disks ,
a detailed exploration of their sed is essential . especially important
are the silicate emission features and the wavelength range between 60 - 180
. the upcoming sirtf and sofia missions will be capable of gathering a statistically relevant number of complete seds , as well as providing a reliable disk frequency .
we present the first comprehensive models of the circumstellar dust surrounding a young bd .
our results are : * the overall shape of the sed is very similar to those of t tauri disks * a spherical dust distribution contradicts the observed extinction * a simple flat disk fits well the observations * the flat geometry hints towards large grains * the classical flared disk does not reproduce the sed of cfht - bd - tau 4 * a flared disk with a puffed - up inner rim can only explain the observed sed if it has an inner gap much larger than the dust sublimation radius c c c c 0.658 & 0.062 & 0.002 & 1 + 0.822 & 1.23 & 0.03 & 1 + 1.25 & 22.9 & 1.0 & 1 + 1.65 & 38.7 & 1.0 & 1 + 2.17 & 46.5 & 1.4 & 1 + 3.75 & 52.5 & 1.5 & 2 + 6.7 & 28.4 & 5.7 & 3 + 14.3 & 38.5 & 7.7 & 3 + 850 & 10.8 & 3.7 & 4 + 1300 & 2.14 & 0.7 & 4 + m1 & rt & shell & 0 & & no & 4&a@xmath1560 mag & ( 1 ) + m2 & rt & shell & @xmath161 & & no & 4&a@xmath15300 mag & ( 1 ) + m3 & rt & shell & @xmath162 & & no & 4&a@xmath151000 mag & ( 1 ) + m4 & aa & flat & one - layer & 0.02 & yes & 1.4 & @xmath17=2045(35 ) & ( 2 ) + m5 & aa & flat & two - layer & 0.01 & yes & 4 & @xmath17=530(10 ) & + m6 & scc & flared&no rim & 0.03 & ( yes ) & 0.7 & @xmath17=80 - 82 & ( 3 ) + m7 & scc & flared&no rim & 0.07 & no & 0.7 & & ( 3 ) + m8 & scc & flared & rim & 0.03 & no & 0.3 & & ( 3 ) + m9 & scc & flared & rim & 0.05 & yes & 0.3&@xmath17=820(12 ) & ( 3 ) + adams , f. c. , lada , c. j. , shu , f. h. 1987 , , 312 , 788 allard , f. , hauschildt , p. h. , alexander , d. r. , tamanai , a. , & ferguson , j. w. 1999 , , volume 212 , 127 apai , d. , pascucci , i. , henning , th . , sterzik , m. f. , klein , r. , semenov , d. , gnther , e. , & stecklum , b. 2002 , 573 , l115 beckwith , s. v. w. , sargent , a. i. , chini , r. s. , & gsten , r. 1990 , , 99 , 924 beckwith , s. v. w. , henning , t. , & nakagawa , y. 2000 , protostars and planets iv , p. 533
briceo c. , luhman , k. l. , hartmann , l. , stauffer , j. r. , & kirkpatrick , j. d. , 580 , 317 calvet , n. , dalessio , p. , hartmann , l. , wilner , d. , walsh , a. , & sitko , m. 2002 , , 568 , 1008 chiang , e. i. , & goldreich , p. 1997
, , 490 , 368 comern , f. , neuhuser , r. , & kaas , a. a. 2000 , , 359 , 269 draine , b. t. , & lee , h. m. 1984 , , 285 , 89 dullemond , c. p. , dominik , c. , & natta , a. 2001 , , 560 , 957 itoh , y. , tamura , m. , & tokunaga , a. t. 2002 , pasj , 54 , 561 jayawardhana , r. , mohanty , s. , & basri , g. 2003 astro.ph 3565j kenyon , s. j. , & hartmann , l. 1987 , , 323 , 714 kirkpatrick , j. d. , henry , t. j. , & irwin , m. j. 1997 , , 113 , 1421 klein , r. , apai , d. , pascucci , i. , henning , th . , & waters , l. b. f. m. 2003 , , submitted knigl , a. 1991 , , 370 , l39 liu , m. c. , najita , j. , & tokunaga , a. t. 2003 , , 585 , 372 manske , v. , henning , th . , & menshchikov , a. b. 1998 , , 331 , 52 martn , e. l. , dougados , c. , magnier , e. , mnard , f. , magazz , a. , cuillandre , j .- c . , & delfoss , x. 2001 , , 561 , l195 mokler , f. , & stelzer , b. 2001 , , 391 , 1025 moraux , e. , bouvier , j. , stauffer , j. r. , cuillandre , j .- c .
2003 , , 400 , 891 muench , a. a. , alves , j. , lada , c. j. , & lada , e. a. 2001 , , 558 , l51 natta , a. , & testi , l. 2001 , , 376 , l22 natta , a. , testi , l. , comern , f. , oliva , e. , dantona , f. , baffa , c. , comoretto , g. , & gennari , s. 2002 , , 393 , 597 phan - bao , n. , et al .
2003 , , 401 , 959 persi , p. et al .
2000 , , 357 , 219 reipurth , g. , & clarke , c. 2001 , , 122 , 432 terebey , s. , chandler , c. j. , & andr , p. 1993
, , 414 , 759 testi , l. , natta , a. , oliva , e. , dantona , f. , comern , f. , baffa , c. , comoretto , g. , & gennari , s. 2001 , , 571 , l155 | the combination of mid - infrared and recent submm / mm measurements allows us to set up the first comprehensive spectral energy distribution ( sed ) of the circumstellar material around a young brown dwarf .
simple arguments suggest that the dust is distributed in the form of a disk .
we compare basic models to explore the disk parameters .
the modeling shows that a flat disk geometry fits well the observations .
a flared disk explains the sed _ only _ if it has a puffed - up inner rim and an inner gap much larger than the dust sublimation radius .
similarities and differences with disks around t tauri stars are discussed . |
hera collides an electron or positron beam of 27.5 gev and a proton beam of 920 gev ( 820 gev until 1997 ) , yielding 318 gev ( 300 gev ) center - of - mass energy ( @xmath4 ) .
the square of the momentum transfer ( @xmath5 ) in deep inelastic scattering ( dis ) can reach several times @xmath6 , which means that the structure of the proton is probed with a wavelength as small as one thousandth ( @xmath7 cm ) of its radius . during the years 1994 - 2000 , each of the two collider experiments , h1 and
zeus , has collected approximately 110 @xmath8 of @xmath9 and 15 @xmath8 of @xmath10 data . with these data ,
the experiments are sensitive to rare processes with a cross section of a fraction of a picobarn , and searches for exotic signals of physics beyond the standard model ( sm ) are extensively performed .
possible signals of new physics include : * resonant states formed by electron - quark fusion . examples of such states are leptoquarks or squarks ( superpartners of quarks ) in supersymmetry ( susy ) with @xmath0-parity violation .
on - shell production of such particles is possible for those with masses up to @xmath4 . *
physics at a much higher energy scale than the hera energy could cause a virtual effect in electron - proton scattering in the highest-@xmath5 domain .
such an effect is modeled as an effective four - fermion ( @xmath11 ) contact interaction .
a variant of such a phenomenon is a lepton - flavor - violating ( lfv ) interaction in which the incoming electron changes its flavor to one of the higher - generation leptons ( @xmath12 or @xmath13 ) . *
if the fermions ( leptons and/or quarks ) are composite rather than elementary , excited states of fermions could be produced as a consequence of the large momentum transfer , if the masses of such states are below the hera @xmath4 .
such excited fermions could decay back to the ground state ( normal fermions ) `` radiatively '' by emitting a gauge boson ( @xmath3 , @xmath14 , @xmath15 or gluon ) . in the following sections ,
recent results are presented from both collaborations on these topics .
most of the results are preliminary .
leptoquarks ( lqs ) are particles which carry both lepton ( l ) and baryon ( b ) numbers .
they appear in many unifying theories which try to establish fundamental relations between leptons and quarks .
if lqs with a mass below the @xmath4 of hera exist , they could be directly produced via a fusion between the electron and a quark in the proton .
the production cross section depends on the unknown yukawa coupling @xmath16 at the lq - electron - quark vertex . to evaluate the experimental results , the theoretical framework of buchmller - rckl - wyler ( brw)@xcite is often used . in this framework , lqs decay to an electron - quark pair with a fixed branching ratio of either 1/2 or 1 , depending on the lq species .
the only other possible decay mode is into a neutrino - quark pair .
when the lq decays to an electron - quark pair , the final state is identical to sm neutral current ( nc ) dis .
however , the angular distribution is different .
nc dis with @xmath1-channel photon exchange has a @xmath17 angular distribution , where @xmath18 is the inelasticity variable related to the electron decay angle in the @xmath19-@xmath20 rest frame , @xmath21 , with @xmath22 . on the other hand , scalar lqs have flat @xmath18 distribution and vector lqs have @xmath23 dependence .
therefore , a cut in @xmath18 ( or @xmath24 ) is used to enhance the lq signal among the nc dis background .
figure [ h1lq ] shows the mass distribution of nc dis candidates from the h1 experiment .
there was an excess of events after a @xmath18 cut in the old @xmath9 data@xcite , which is not confirmed by the new @xmath9 data taken at higher @xmath4@xcite .
zeus also had a small excess in the high-@xmath25 , high-@xmath18 region ( @xmath25 is the bjorken scaling variable , related to the mass of the @xmath26 system via @xmath27=@xmath28 ) in the old data , which also is not confirmed by the new data@xcite . with the smaller amount of @xmath10 data already analyzed ,
both experiments have observed good agreement with sm predictions@xcite . with no evidence of a signal ,
limits are derived on the yukawa coupling as a function of the lq mass , as shown in fig .
[ feq2lim ] .
note that @xmath9 and @xmath10 data have complementary sensitivity to different lq species ; @xmath9 collisions mainly probe the production of @xmath29=0 lqs , which couple to a positron and a valence quark ( @xmath29 is the fermion number , defined as @xmath29=@xmath30 + 3@xmath31 ) , and correspondingly @xmath10 for @xmath29=2 . in the figure ,
comparison is made with other collider results . at the tevatron
, lqs can be pair - produced and thus the limit on mass is independent of @xmath16 .
the limits from lep come from hadronic cross section measurements in which lq exchange can produce virtual effects .
one can also consider a general case and treat the lq branching ratio to @xmath26 as a free parameter , in contrast to the restrictions of the brw framework . in fig .
[ zeusnujet ] , the search results in both nc ( electron - jet resonance ) and cc ( neutrino - jet resonance ) channels are used , and combined limits are obtained assuming the branching ratios to @xmath26 and @xmath32 decays sum up to one@xcite .
the limits in lq mass for fixed values of @xmath16 are largely independent of how the branching ratios are shared , while the tevatron limits degrade largely when the lq predominantly decays to @xmath32 .
@xmath0-parity is a multiplicative quantum number defined as @xmath33 , where @xmath34 is the spin of the particle .
it takes the value @xmath35 for usual particles and @xmath36 for their superpartners . in the minimal supersymmetric standard model ( mssm ) , @xmath0-parity
is assumed to be conserved .
the consequences are that superparticles are always produced in pairs , and that the lightest superparticle is stable .
however , the general supersymmetric and gauge - invariant superpotential contains three terms which violate @xmath0-parity , two with @xmath30 violation and one with @xmath31 violation@xcite .
of particular interest at hera is the term @xmath37 , where @xmath30 and @xmath38 are left - handed lepton and quark doublet superfields and @xmath39 is a right - handed singlet superfield of down - type quarks .
the indices @xmath40 , @xmath41 and @xmath42 denote their generation . for each generation combination ,
a new yukawa coupling @xmath43 is introduced . with a presence of non - zero @xmath44 , production of squarks
is possible in @xmath26 fusion at hera , in exact analogy to the scalar lq production described before .
however , the decay mode of the squark has more varieties than lq decays in the brw framework .
in addition to the @xmath26 decay , a squark can decay to a quark and a gaugino with @xmath0-conserving gauge couplings .
the gauginos can decay to a lepton ( @xmath19 or @xmath45 ) and two quarks with the same @xmath43 coupling , or to a lighter gaugino and two fermions with gauge couplings which results in further cascade decays . as a consequence , there are many multi - jet and/or multi - lepton final states with possible missing momentum , which compete with the simple lq - like decays .
it is worth noting that both @xmath46 and @xmath47 final states are possible in the gaugino decay to @xmath48 , which makes possible a `` wrong sign '' electron search with respect to the charge of the beam electron , a channel with very little background from sm processes .
both h1 and zeus experiments have analyzed the @xmath9 data taken during 1994 - 97 ( approx .
40 @xmath8 per experiment ) in many decay channels , and no evidence of squark production was found .
the results can then be converted to constraints in susy parameter space under various scenarios .
figure [ h1zeussusy ] ( left ) shows the zeus results in the unconstrained mssm framework , in which squark masses are treated as parameters independent of the mssm parameters @xmath12 , @xmath49 and @xmath50 . for two fixed values of @xmath50 , a scan in the large parameter space of @xmath12 and
@xmath49 has been performed , and upper limits in @xmath51 have been obtained as a function of the squark mass@xcite . the h1 collaboration has also derived limits in the msugra framework , where the sparticle masses are determined from other susy parameters and the number of free parameters become minimal@xcite .
the results are presented in fig .
[ h1zeussusy ] ( right ) . for fixed values of @xmath43 and @xmath50 , the excluded region in susy parameter space ( @xmath52 , @xmath53 )
for reasonably large values of the yukawa coupling ( even as small as 0.1 ) , hera has a sensitivity in a region not explored at the tevatron , although it should be mentioned that the tevatron limits are independent of @xmath43 .
new physics at an energy scale which is much larger than the @xmath4 of hera could also be detected via its virtual effects at the highest-@xmath5 region ( @xmath54 several @xmath55 gev@xmath56 ) which corresponds to an @xmath19-@xmath20 interaction at a distance as short as 0.001 fm .
such new physics processes are modeled in their low - energy limit as an effective four - fermion contact interaction ( ci ) , and the effects are parameterized by a coupling constant , @xmath57 , and an effective mass scale , @xmath58 .
both collaborations@xcite tested the most common @xmath11 ci models using the nc dis data at high @xmath5 .
since the data are well described by the sm within statistical errors , limits are derived for each model tested .
the limit on @xmath58 depends on the chiral structure of the model , and ranges up to 9 tev when taking the usual coupling convention @xmath59=@xmath60 .
the hera limits are competitive and complementary to those from lep and the tevatron , where the tests are made in @xmath61hadrons and drell - yan production , respectively .
limits are also set on some specific models . in models with large ( compactified ) extra dimensions in which only the gravitational force can propagate to the extra dimensions
, the fundamental planck mass scale could be as low as the electroweak scale ( @xmath62 tev)@xcite .
in such models , high - energy particle collisions get additional contributions due to the exchange of a kaluza - klein tower of gravitons which modify the cross sections at the highest energies .
this causes ci - like effects at hera in the highest-@xmath5 region with a main contribution from the interference of graviton exchange and normal nc exchange .
these effects are characterized by an effective coupling @xmath63 , where @xmath64 is close to the fundamental planck scale and @xmath65 determines whether the interference is constructive or destructive .
the results from h1@xcite are shown in fig .
[ h1ci ] ( left ) .
an effective mass scale @xmath64 of nearly 1 tev is excluded at 95% cl .
another example is a test of classical form factor of quarks .
if one assumes that quark has a finite size and its charge is distributed with a r.m.s . of @xmath0 ,
the nc dis cross section gets a suppression factor @xmath66 which dumps the cross section as @xmath5 increases .
radius of the order of @xmath7 cm is excluded , as shown in fig .
[ h1ci ] ( right)@xcite .
table 1 : _ zeus limits ( 95% cl upper limit ) on @xmath67 @xmath68 for @xmath29=0 lfv leptoquarks mediating the @xmath69 transition ( bold numbers in the bottom of each cell ) .
each row corresponds to a @xmath70 generation combination and each column corresponds to a lq species .
the numbers in the middle of each cell are the best limit from other experiments .
the cases where the zeus limit is the most stringent or comparable to the best limit within a factor of 2 are enclosed in a box .
the * shows the cases where only the top quark can participate .
similar tables exist for @xmath29=2 lqs , and for the @xmath19-@xmath12 transition . _ 0.1 in in the sm
, lepton flavor is conserved in every interaction .
however , it is not regarded as a fundamental symmetry , and new physics could naturally entail the inter - generation transition of leptons .
hera is well suited for searching for electron - muon or electron - tau lfv processes by colliding an electron with a proton at a very short distance .
the final state will be an @xmath71 event which has a striking signature of a high - momentum muon or tau balanced by a hadronic system , which has very little background from sm processes . both experiments searched for such events in the 1994 - 97 data , with no candidates@xcite .
an example of the limits are expressed in table 1 in the context of heavy ( @xmath72 ) lfv leptoquarks mediating the @xmath73 transition , for each lq species and @xmath74 generation combination .
it can be seen that limits from low - energy experiments , such as rare decays , are very stringent for transitions involving first - generation quarks , but hera has higher ( or unique ) sensitivity when the lfv transition involves higher - generation quarks .
if fermions are not elementary but have substructure , their excited states could be produced in a high - energy collision . at hera
, excited electrons and quarks could be produced mainly via @xmath1-channel photon exchange , and excited neutrinos via @xmath14 exchange .
once produced , they can decay to a normal fermion by radiating off gauge bosons .
decays into @xmath3 , @xmath14 or @xmath15 are considered here . to measure the experimental sensitivity , the theoretical framework of hagiwara , komamiya and zeppenfeld@xcite
is conventionally used , in which the production cross section depends on the relative coupling strengths @xmath75 , @xmath76 and @xmath77 for the su(2 ) , u(1 ) and su(3 ) gauge groups , respectively , and a common compositeness scale @xmath58 . by assuming relations between the couplings ,
the decay branching ratios to different gauge bosons are fixed and the cross section depends only on the parameter @xmath78 .
both collaborations searched for excited fermions in various decay modes ( including hadronic and leptonic decays of @xmath14 , @xmath15 ) in the 1994 - 97 @xmath9 data and 1998 - 99 @xmath10 data , with no evidence of a signal@xcite .
it should be noted that @xmath10 collisions provide a much larger cross section for excited - neutrino production than @xmath9 because of the larger @xmath2-quark density in the proton and the helicity suppression in @xmath9 due to @xmath14-exchange .
figure [ exf ] shows the upper limits obtained on @xmath78 as a function of the excited - fermion mass , for @xmath79 and @xmath80 .
hera experiments have a unique sensitivity beyond the lep2 center - of - mass energy .
stringent limits on @xmath80 are obtained from the tevatron , but the production at the tevatron takes place via the strong coupling @xmath77 .
hera limits , on the other hand , are sensitive to the electroweak couplings @xmath75 and @xmath76 and are thus complementary .
0.1 in . _ summary of isolated high-@xmath81 leptons from h1 ( top ) and zeus ( bottom ) . for two values of cuts in @xmath82 ,
the observed and expected numbers of events are shown .
the numbers in parenthesis are the w contribution to the total sm expectation . for h1 , the error in the expected number includes the systematic error . _ [ cols="^,^,^ " , ] [ pttab ] events with a high transverse momentum ( high-@xmath81 ) lepton and missing @xmath81 at hera have drawn attention in recent years due to the excess of such events reported by the h1 experiment . when the region with large @xmath82 ( the transverse momentum of the remaining hadronic system ) is selected , the prediction from the sm is dominated by single-@xmath14 production , which has a total cross section of about 1 pb . figure [ h1hpt ] shows the results from h1 presented in summer 2000 , from the @xmath9 data collected up to middle of 2000@xcite .
events are plotted in @xmath82 and @xmath83 , the transverse mass between the lepton and missing @xmath81 .
there is no acceptance for muon events at low @xmath82 due to the requirement of missing @xmath81 measured in the calorimeter .
the observed and expected numbers of events are shown in table [ pttab ] ( top ) . for @xmath84 gev , nine electron or muon events were observed while @xmath85 are expected from the sm , dominated by @xmath14 production .
zeus had observed good agreement with sm predictions for these event topologies up to the results presented in summer 2000@xcite , which included data from 1994 - 99 . in this conference , the results were updated for the first time using all of the data collected in 2000 .
0.3 cm figures [ zeusptx ] and [ zeusmt ] show the distributions of @xmath82 and transverse mass for events with calorimeter @xmath86 gev and an isolated track with @xmath87 gev .
ten electron events and seven muon events were selected , with the sm expectation being @xmath88 and @xmath89 , respectively .
both the event rate and the distribution of the kinematic variables are in good agreement with sm expectations , which are dominated by nc dis for electrons and @xmath90 for muons . by further applying cuts to enhance the @xmath14 component in the expectation , similar to the ones used in the h1 analysis ,
the numbers of events are compared with the sm predictions in table [ pttab ] ( bottom ) for the large-@xmath82 region . in the zeus case , there is no excess of events observed .
there is somewhat larger acceptance for @xmath14 in the h1 analysis due to the wider polar - angle coverage , but it is worth mentioning that all of the observed h1 events are in the polar - angle range where zeus has acceptance .
the muon event surviving the @xmath84 gev cut is from the 2000 data , and its event display is shown in fig .
[ nicemuon ] .
an event with a lepton , missing @xmath81 and a large @xmath82 is a typical signature for a top quark decaying via @xmath91 .
therefore , the results of the searches described above can be applied to single - top production , @xmath92 .
it is a flavor - changing neutral current ( fcnc ) process and highly suppressed in the sm .
any observation of single top production at hera would be a signal for physics beyond the sm .
the resulting limits is also used . ] , together with those from lep@xcite and the tevatron , are expressed in fig .
[ fcnc ] in the plane of two couplings , @xmath93 , the magnetic coupling at the photon - top - quark vertex and @xmath94 , the vector coupling at the @xmath15-top - quark vertex . here
the quark can be generally @xmath2 or @xmath95 . at lep ,
the limits come from a single - top search @xmath96 and the tevatron limits are from a rare top decay search @xmath97 .
both processes can constrain the @xmath3 and @xmath15 couplings , and the limits apply to both @xmath2 and @xmath95 quarks since they are in the final state . at hera , the contribution from @xmath15-exchange in the @xmath1-channel is suppressed due to the large mass in the propagator , so the cross section is dominated by the photon coupling @xmath93 . moreover , since the @xmath2 quark dominates the parton distribution in the proton at large @xmath25 , hera is most sensitive to the @xmath98 coupling . it can be seen that the hera limits on @xmath99 are more stringent than those from lep and the tevatron .
note that the cross section dependence on the top mass uncertainty ( 5 gev ) is about 20% at hera@xcite , while it is larger at lep where the available energy is close to the threshold .
during the `` hera 1 '' running , both h1 and zeus experiments collected about 110 pb@xmath100 of @xmath9 and 15 pb@xmath100 of @xmath10 data .
not all of the searches have finished analyzing all of the data yet , but so far no evidence for new physics has been observed .
therefore , new constraints on various physics have been derived : leptoquarks , squarks in @xmath0-parity violating susy , @xmath11 contact interactions , large extra dimensions , quark radius , lepton flavor violation and excited fermions .
limits are comparable and largely complementary to those obtained in searches at lep or the tevatron .
the excess of isolated leptons in the h1 data is intriguing , although the zeus results from the total data sample are consistent with sm expectations .
limits on single - top production yield the most stringent constraint on the fcnc coupling at the @xmath98 vertex .
hera and both experiments have been performing shutdown work since september 2000 for the luminosity upgrade .
new focusing magnets are installed very close to the interaction point , even inside the detectors .
the new optics , together with a moderate increase in beam currents , will bring a five - fold increase in the instantaneous luminosity compared to the year 2000 running .
in addition the detectors will be upgraded ; for example , zeus will have a micro - vertex detector for the first time . from summer 2001 , the machine will restart and physics running is scheduled from december . in the five - year program of `` hera 2 '' running , approximately 1 fb@xmath100 of data are expected per experiment .
another notable feature of the new running is that the longitudinal polarization of the electron / positron beam will be available by default to the collider experiments . in these five years ,
hera 2 and tevatron run ii will be the only energy - frontier machines in operation
. they will compete in many physics scenarios .
for example , leptoquark ( or squark ) sensitivity is illustrated in fig .
[ perez]@xcite .
if the yukawa coupling @xmath16 is within the sensitivity of hera , and if hera 2 quickly starts up with high - luminosity running , there is a chance for discovery with 10 times more data to come . otherwise , for the case of leptoquarks decaying 100% to @xmath26 , with 1 - 2 fb@xmath100 of run ii data the tevatron experiments can close the discovery window for direct observation at hera since the @xmath16-independent mass limit will exceed the hera kinematic limit .
however , there are cases where hera has still potential not covered by run ii ; as indicated in the figure , in the case of leptoquarks with a small branching ratio to @xmath26 , hera s sensitivity exceeds that of run ii even for fairly small yukawa coupling . also there are some models
not probed extensively at the tevatron such as excited leptons .
therefore , the two colliders will continue to be competitive and complementary during the coming years before the start of lhc , in the hunt for possible excitement .
the author is very grateful to the organizers of `` xv rencontres de physique de la vallee daoste '' for their excellent hospitality which ensured a pleasant stay at la thuile .
99 w. buchmller , r. rckl and d. wyler , phys . lett . *
b191 * , 442 ( 1987 ) .
_ ibid . _ * b448 * , 320 ( 1999 ) .
h1 collaboration , eur .
j. * c11 * , 447 ( 1999 ) .
_ ibid . _ * c14 * , 553 ( 2000 ) .
h1 collaboration , abstract 953 submitted to ichep00 , osaka .
zeus collaboration , eur .
j. * c16 * , 253 ( 2000 ) . +
zeus collaboration , abstract 1038 submitted to ichep00 , osaka .
zeus collaboration , abstract 552 submitted to eps hep99 , tampere .
+ h1 collaboration , abstract 954 submitted to ichep00 , osaka .
zeus collaboration , phys . rev . *
d63 * , 052002 ( 2001 ) .
v. barger , g.f .
giudice and t. han , phys .
* d40 * , 2987 ( 1989 ) .
zeus collaboration , abstract 1042 submitted to ichep00 , osaka .
h1 collaboration , desy 01 - 021 , submitted to eur .
j. zeus collaboration , eur .
j. * c14 * , 239 ( 2000 ) .
h1 collaboration , phys .
b479 * , 358 ( 2000 ) .
+ h1 collaboration , abstract 951 submitted to ichep00 , osaka .
n. arkani - hamed , s. dimopoulos and g. dvali , phys . lett . * b429 * , 263 ( 1998 ) .
_ , phys . rev . *
d59 * , 086004 ( 1999 ) .
zeus collaboration , abstract 1039 submitted to ichep00 , osaka .
k. hagiwara , s. komamiya and d. zeppenfeld , z. phys .
* c29 * , 115 ( 1985 ) .
h1 collaboration , eur .
j. * c17 * , 567 ( 2000 ) .
+ h1 collaboration , abstract 956 submitted to ichep00 , osaka . + zeus collaboration , abstract 555 submitted to eps hep99 , osaka . + zeus collaboration , abstract 1040 submitted to ichep00 , osaka .
h1 collaboration , abstract 976 submitted to ichep00 , osaka .
+ h1 collaboration , abstract 961 submitted to ichep00 , osaka .
zeus collaboration , abstract 1041 submitted to ichep00 , osaka .
aleph collaboration , phys . lett .
* b494 * , 33 ( 2000 ) .
a. belyaev and n. kidonakis , hep - ph/0102072 .
m. kuze , s. lola , e. perez and b. allanach , hep - ph/0007282 . | recent results from searches for new physics at hera are reviewed . exploiting the uniqueness of lepton - hadron collisions at high energy ,
searches are performed for electron - quark resonant states ( leptoquarks or squarks in @xmath0-parity - violating supersymmetry ) or excited states of fermions .
new phenomena at a very high energy scale , manifested at present energies as effective four - fermion contact interaction , are also investigated , including cases with lepton - flavor violation .
finally , the status of events with a high - energy lepton and missing transverse momentum is presented , resulting in the most stringent constraint on the flavor - changing neutral current @xmath1-@xmath2-@xmath3 coupling which could yield single - top production .
= 14.5pt |
bose - einstein condensation , first predicted in the years 1924 - 25 @xcite , was observed in 1995 on rubidium and sodium vapors @xcite .
these remarkable experiments have generated since then a clear interest in the critical properties of this dilute macroscopic quantum state of matter .
in particular , the transition temperature @xmath1 is a crucial parameter whose exact determination has been a matter of discussion for decades @xcite .
it is indeed rather difficult to estimate the influence of atomic interactions on the condensation temperature for a _ uniform _ dilute weakly interacting bose gas .
this comes from the fact that this is a many - body problem affected by long - range critical fluctuations which have to be described non - perturbatively @xcite . as a consequence ,
many different predictions were produced since the late 1950 s , like an increase or even a decrease of @xmath1 proportional to @xmath13 , to @xmath14 , to @xmath15 or even to @xmath16 , where @xmath14 denotes the atomic @xmath17-wave scattering length @xcite .
it is only in 1999 that this question was settled with the rigorous demonstration of the linear increase of the critical temperature with @xmath14 @xcite .
it was later shown @xcite that the relative correction @xmath18 behaves at second order as @xmath19 , where @xmath20 denotes the condensation temperature of the ideal gas in the thermodynamic limit , and @xmath21 , @xmath22 being the ( uniform ) atomic density . in this expression , @xmath23 , @xmath24 and @xmath25 @xcite .
on the other hand , in a harmonic trap long - range critical fluctuations are suppressed , and the leading ( linear ) order correction to @xmath1 can be calculated by simple perturbative methods @xcite .
using a non - perturbative approach , it was shown in 2001 that in such a potential the relative correction @xmath18 behaves also at second order as @xmath26 , with @xmath27 , @xmath28 being the thermal de broglie wavelength @xcite . in this case ,
@xmath29 , @xmath30 and @xmath31 @xcite .
these results were then extended to power - law potentials by zobay _
et al _ in a series of beautiful papers first using mean - field theory @xcite , then using renormalization group theory @xcite , and later on using variational perturbation theory @xcite .
these different theoretical studies helped bridging the gap separating _ homogeneous _ from _ inhomogeneous _ potentials by showing how an increasingly inhomogeneous potential suppresses the influence of critical fluctuations . in these studies , the shape of the potential
could be summarized by a simple parameter @xmath32 which varies between @xmath33 for the uniform gas ( homogeneous case ) and 2 for the ( inhomogeneous ) harmonic trap . in ref .
@xcite , it was shown for instance that critical fluctuations have a marginal impact on the transition temperature when @xmath34 for @xmath35 . the amplitude of the corrections due to atomic interactions increases of course with the number of particles . on the other side , independently of atomic interactions , with a finite number of particles
the transition temperature is different from the one obtained in the thermodynamic limit . in the case of a harmonic trap ,
this finite size effect results in a downward shift of the transition temperature , which is , at lowest order , linear in the zero point energy of the trapping potential @xcite . [ cols="<,^,^,^,^,^,^ " , ] [ tab : tc ] one can first notice in table [ tab : tc ] that the case @xmath36 is a special case .
indeed , for @xmath36 all finite atom number corrections cancel out in eq .
( [ eq : real ] ) except for the constant term @xmath37 .
@xmath1 is then identical to @xmath20 for all @xmath0 .
this cancellation comes from the fact that incidentally @xmath38 and @xmath39 [ see eqs .
( [ eq : abc ] ) ] .
this is the only value of the shape parameter @xmath32 where there is rigorously no finite atom number correction , whatever the number of trapped atoms .
this special case corresponds for instance to an isotropic cubic trap , with @xmath8 , but also to a trapping potential of cylindrical symmetry such as @xmath40 or to @xmath41 in the case of a cartesian power - law trap .
it also applies for instance to @xmath42 , or to any combination of power - law potentials for which @xmath43 [ see eq .
( [ eq : eta ] ) ] .
the variation of the relative correction @xmath18 is shown as a function of the shape parameter @xmath32 in fig .
[ fig : dtc - eta ] for @xmath44 ( solid line with black circles ) and @xmath45 ( dashed line with red squares ) .
from the magnified inset of this figure , one can notice again that the point @xmath36 plays a particular role since trapping potentials with @xmath46 are characterized by a decrease of @xmath1 due to finite size effects , while @xmath47 leads , on the contrary , to an increase of @xmath1 .
this could already be inferred from table [ tab : tc ] , where it is seen that the lowest order correction is positive when @xmath47 and negative when @xmath46 .
in addition , one can notice in fig .
[ fig : dtc - eta ] that , even when @xmath48 , the increase in @xmath1 becomes significantly large when @xmath49 .
indeed , for @xmath50 , an increase of 37% is expected when @xmath51 , and an increase of 84% is expected when @xmath52 .
a large upward shift of @xmath1 can therefore be expected in the case of a generic high - order power - law potential , like @xmath9 with @xmath10 for instance .
one should note here that we have limited our study to @xmath53 , _
i.e. _ to inhomogeneous potentials since a non - perturbative treatment would be necessary for an accurate estimation of @xmath54 in the homogeneous limit @xmath55 .
( in % ) due to finite size effects as a function of the shape parameter @xmath32 for @xmath56 ( solid line with black circles ) and @xmath57 ( dashed line with red squares ) .
the inset is a simple magnification of the region surrounding @xmath36 .
note that @xmath58 for @xmath36 , @xmath59.,width=325 ] in order to check the influence of these corrections in a realistic trapping situation , we now turn to the case of a rubidium gas trapped by blue - detuned laguerre - gaussian optical beams . +
recently , we have proposed an original route to achieve bose - einstein condensation using dark power - law laser traps @xcite .
experimentally , it has been demonstrated that cold atoms can be trapped and guided in such optical setups @xcite .
these traps are created with two crossing blue - detuned laguerre - gaussian optical beams .
a simple control of their azimuthal order @xmath60 allows for the exploration of a wide range of situations , with atoms trapped in confining power - law potentials in one , two , and three dimensions .
it was shown that , for a fixed atom number , higher transition temperatures are obtained in configurations where @xmath61 , compared to a harmonic trap @xmath62 of the same size .
we investigate here finite size effects in such optical traps . to create a three - dimensional trap
, we consider an all - optical configuration consisting of two perpendicularly crossing laguerre - gauss laser modes of same radial index @xmath63 and azimuthal index @xmath60 .
the polarizations of the two beams are chosen to be orthogonal to avoid any interference pattern .
the first beam , circularly symmetric , propagates along the @xmath64 direction and provides trapping in the @xmath65 plane . trapping in the third dimension
is provided by another strongly elongated laguerre - gaussian beam shaped elliptically in the form of a light sheet and propagating in the @xmath66 direction .
if the two beams are characterized by the same laser power , waist and detuning , the corresponding potential near the trap center can be described by @xmath67 @xmath68 is then a simple function of the laser power , detuning and waist of the two beams @xcite , and @xmath69 .
( in % ) due to finite size effects as a function of the shape parameter @xmath32 ( lower scale ) and of the azimuthal index @xmath60 ( upper scale ) for two laguerre - gaussian beams of power @xmath70w and detuning @xmath71thz . the solid black line with circles corresponds to @xmath72 particles , the red dashed line with squares to @xmath73 and the blue dash - dotted line with triangles to @xmath74 .
the volume of the trap has been chosen such that , for each atom number , the condensation temperature @xmath20 obtained in the thermodynamic limit for @xmath75 is 100nk ( see @xcite for details).,width=325 ] the relative correction @xmath18 due to finite size effects in such optical traps is shown in fig .
[ fig : dtc - l ] as a function of the shape parameter @xmath32 ( lower scale ) or equivalently of the azimuthal index @xmath60 ( upper scale ) , for realistic values of all optical parameters ( see figure caption ) .
one can notice in this figure that , even for large condensate occupation numbers , the shift of the condensation temperature due to finite size effects is not negligible when @xmath76 .
indeed , for @xmath77 this correction is of about 60% , 38% and 16% for @xmath78 , @xmath11 and @xmath79 , respectively .
this potential corresponds to a 12@xmath80-power power - law trap , which can be obtained practically using @xmath81 laguerre - gaussian beams @xcite .
( lower scale ) and of the azimuthal index @xmath60 ( upper scale ) for two laguerre - gaussian beams of power @xmath70w and detuning @xmath71thz .
the solid green line with black crosses corresponds to the condensation temperature @xmath20 obtained in the thermodynamic limit .
the dashed black line with circles gives the condensation temperature @xmath1 for @xmath72 particles , the red dashed line with squares for @xmath73 and the blue dashed line with triangles for @xmath74 .
the volume of the trap has been chosen such that , for each atom number , the condensation temperature obtained in the thermodynamic limit for @xmath75 is 100nk ( see @xcite for details).,width=325 ] similarly , fig .
[ fig : tc - l ] shows the variation of the condensation temperature with the shape parameter @xmath32 and the azimuthal index @xmath60 .
the temperature @xmath20 obtained in the thermodynamic limit is shown as a thick green solid line .
it is identical for all atom numbers @xmath0 since we have chosen the volume of the trap such that the condensation temperature obtained in the thermodynamic limit is the same for all atom numbers ( 100nk with @xmath75 , see @xcite for details ) .
as already discussed in @xcite , the temperature @xmath20 increases with @xmath60 since @xmath82 [ see eq .
( [ eq : tc0 ] ) ] . in the same figure ,
the condensation temperature @xmath1 taking into account finite size effects is shown as a dashed black line with circles for @xmath72 , a red dashed line with squares for @xmath83 and a blue dashed line with triangles for @xmath84 .
compared to a harmonic trap of the same size , it can be concluded that much higher critical temperatures can be obtained in a power - law trap , and this especially for small condensation occupation numbers .
this large increase results from the combination of two positive effects : first the effect of the potential shape on the temperature @xmath20 [ eq .
( [ eq : tc0 ] ) ] and second the finite size effect [ eq .
( [ eq : comptc ] ) ] . indeed , with the optical parameters chosen in fig .
[ fig : tc - l ] , compared to @xmath85nk in the harmonic case ( @xmath75 ) , the condensation temperature for @xmath77 is about 283nk , 251nk and 212nk for @xmath78 , @xmath11 and @xmath79 , respectively.this corresponds to an increase of @xmath1 by about 183% , 151% and 112% , respectively , when compared to the transition temperature found in a harmonic trap .
in this paper , we have presented a detailed theoretical analysis of finite size effects on the transition temperature @xmath1 of an ideal bose gas trapped in an inhomogeneous generic three - dimensional power - law potential . using the local density approximation ,
we have derived an analytical expression of @xmath1 , expressed in terms of a power series in @xmath2 , where @xmath3 denotes the zero - point energy of the trapping potential .
we have also given numerical expressions truncated at second order for a specific set of power - law potentials .
more precisely , we have shown that in the case of a harmonic trap the usual lowest - order correction is inaccurate when the number particles @xmath0 is less than @xmath86 . in this case
, the next order has also to be taken into account , yielding to the relative correction @xmath87 where @xmath88 and @xmath89 denote the arithmetic and geometric means of the harmonic oscillator frequencies . beyond the specific case of the harmonic potential , we have also shown that , if the atoms are trapped in a power - law potential characterized by a shape parameter @xmath36 , all finite size effects cancel out , and the critical temperature is simply equal to the one obtained in the thermodynamic limit , independently of the number of particles . in the case of a trap of cylindrical symmetry , this happens for example for @xmath40 , but also for @xmath90 .
in addition , we have shown that in a power - law potential characterized by a shape parameter @xmath91 , the temperature shift due to finite size effects becomes positive .
this shift may be relatively large for @xmath76 with relatively small atom numbers ( @xmath92 ) .
the magnitude of this phenomenon was finally confirmed in the specific case of rubidium atoms confined by blue - detuned laguerre - gauss optical laser beams .
finally , one should note that even though our approach gives the exact result for all types of power - law potentials within the limits imposed by the approximations used in our theoretical model , it is not able to capture the essence of critical phenomena taking place in a purely homogeneous system , due to the importance of long - range non - perturbative critical fluctuations in such systems . the institut des sciences molculaires dorsay ( ismo ) is unit mixte de recherche de luniversit paris - sud 11 et du cnrs ( umr 8214 ) . the authors would like to acknowledge financial support from agence nationale de la recherche ( projects imagefemto no .
anr-07-blan-0162 - 02 and attowave no .
anr-09-blan-0031 - 01 ) and from the rgion ile - de - france ( programme setci ) . | bose - einstein condensation ( bec ) of an ideal gas is investigated , beyond the thermodynamic limit , for a finite number @xmath0 of particles trapped in a generic three - dimensional power - law potential .
we derive an analytical expression for the condensation temperature @xmath1 in terms of a power series in @xmath2 , where @xmath3 denotes the zero - point energy of the trapping potential .
this expression , which applies in cartesian , cylindrical and spherical power - law traps , is given analytically at infinite order .
it is also given numerically for specific potential shapes as an expansion in powers of @xmath4 up to the second order .
we show that , for a harmonic trap , the well known first order shift of the critical temperature @xmath5 is inaccurate when @xmath6 , the next order ( proportional to @xmath7 ) being significant . we also show that finite size effects on the condensation temperature cancel out in a cubic trapping potential , _
e.g. _ @xmath8 .
finally , we show that in a generic power - law potential of higher order , _
e.g. _ @xmath9 with @xmath10 , the shift of the critical temperature becomes positive .
this effect provides a large increase of @xmath1 for relatively small atom numbers .
for instance , an increase of about + 40% is expected with @xmath11 atoms in a @xmath12 trapping potential . |
we are concerned with the existence , uniqueness , and asymptotic behavior of steady transonic flows past a curved wedge , involving transonic shocks , governed by the two - dimensional full euler equations .
when a supersonic flow past a straight - sided wedge whose half - angle is less than the detachment angle , a shock is expected to form , which is attached to the wedge vertex . if the upstream steady flow is a uniform supersonic state , we can find the corresponding constant flow downstream along the straight - sided wedge boundary , together with a straight shock separating the two states ( see fig .
[ fig - perturbedshock ] ) , by using the shock polar determined by the rankine - hugoniot jump conditions and the entropy condition ( _ cf .
[ fig - shockpolaruv ] ) .
however , the downstream state is not uniquely solvable . in general , there are two solutions
, one of which corresponds to a weaker shock than the other .
as normally expected , a physically admissible shock should be stable under small perturbations .
therefore , it is important to analyze the stability of these shocks in order to understand underlying physics .
the wedge problem described above has a long history at least dating back to the 1930s .
prandtl @xcite in 1936 first conjectured that the weaker shock solution is stable , and hence physically admissible .
there has been a long debate about whether the stronger shock is stable for decades ; see courant - friedrichs @xcite , section 123 , and von neumann @xcite .
see also liu @xcite and serre @xcite .
when the downstream flow is supersonic , the corresponding shock is called a supersonic shock , which is a weaker shock .
this case has been analyzed for the potential flow equation in @xcite with certain convexity assumption on the wedge and in @xcite for an almost straight - sided wedge .
the existence and stability of the steady supersonic shocks for the full euler equations under the bv perturbation of both the upstream flow and the slope of the wedge boundary have been established in chen - zhang - zhu @xcite and chen - li @xcite for lipschitz wedges . for transonic shocks ( _ i.e. _ , the downstream flow is subsonic ) , there are two cases : the transonic shock with the subsonic state corresponding to arc @xmath0 ( which is a weaker shock ) and the one corresponding to arc @xmath1 ( which is a stronger shock ) ( see fig .
[ fig - shockpolaruv ] ) .
the strong shock case has been studied in chen - fang @xcite for the potential flow .
plane , height=257 ] it is well known that the jump of the entropy function across the shock is of cubic order of the shock strength . in general ,
the strength of transonic shocks is large , so the full euler system is a more accurate model than the potential flow or isentropic euler equations . in fang
@xcite , the euler equations are studied with a uniform bernoulli constant for both strong and weak transonic shocks . because the framework is the weighted sobolev space ,
the asymptotic behavior of the shock slope or subsonic solution was not derived . on the other hand
, the asymptotic behavior can be seen more conveniently within the framework of hlder spaces . in yin - zhou @xcite ,
the hlder norms were used for the estimates of the full euler equations with the assumption on the sharpness of the wedge angle , which means that the subsonic state is near point @xmath2 in the shock polar . in chen - chen - feldman @xcite , the weaker transonic shock , which corresponds to the whole arc @xmath0 ,
was investigated ; and the existence , uniqueness , stability , and asymptotic behavior of subsonic solutions were obtained . in @xcite , a potential function is used to reduce the four euler equations into one elliptic equation in the subsonic region .
the method was first proposed in @xcite and has the advantage of integrating the conservation properties of the euler system into a single elliptic equation . however , working on the potential function further requires its lipschitz estimate , besides the @xmath3estimate , to keep the subsonicity of the flow .
there are other related papers about transonic shocks , such as @xcite for transonic flows past three - dimensional wedges and @xcite about transonic flows past a perturbed cone ; see also @xcite for the approaches developed earlier for dealing with transonic shock flows .
the purpose of this paper is to analyze both strong and weak transonic shocks and establish the existence , uniqueness , and asymptotic behavior of the subsonic solutions under the perturbation of both the upstream supersonic flows and the wedge boundaries . through rigorous mathematical analysis
, we are able to prove the stability of both weak and strong transonic shocks .
the strategy is to use the physical variables to make the estimates , instead of the potential function .
the advantage of this method is that only the lower regularity ( _ i.e. _ , the @xmath3estimate ) is sufficient to guarantee the subsonicity .
furthermore , estimating the physical state function @xmath4 directly ( see equations ) also yields a better asymptotic decay rate : for weaker transonic shocks , the decay rate is only @xmath5 in our earlier paper @xcite ; while , in this paper , we will show that the subsonic solution decays to a limit state at rate , @xmath6 , with @xmath7 depending only on the background states ( see remark [ rem:2.2a ] ) .
more precisely , we first use the lagrangian coordinates to straighten the streamlines .
the reason for this is that the bernoulli variable and entropy are conserved along the streamlines , and using the streamline as one of the coordinates simplifies the formulation , especially for the asymptotic behavior of the solution .
then , as in @xcite , we decompose the euler system into two algebraic equations and two elliptic equations .
differentiating the two elliptic equations gives rise to a second - order elliptic equation in divergence form for the flow direction @xmath8 .
given @xmath9 in an expected function space for solutions , we can solve for a new variable @xmath10 as the solution of the linear equation for iterations whose coefficients are evaluated on the given function @xmath9 . once we solve for @xmath10 and obtain the desired estimates , the rest variables are then updated .
thus , we construct a map @xmath11 , where @xmath12 and @xmath13 are the perturbations from the background subsonic state .
the estimates based on our method do not yield the contraction for @xmath14 .
therefore , the banach fixed point argument does not work ; instead , we employ the schauder fixed point argument to obtain the existence of the subsonic solution . for the uniqueness , we estimate the difference of two solutions by using the weighted hlder norms with a lower decay rate .
one point we want to emphasize here is that the decay pattern is different from that for potential flow . in a potential flow ,
the decay is with respect to @xmath15 .
for example , if @xmath16 converges to @xmath17 at rate @xmath5 , then @xmath18 converges at rate @xmath6 . for the euler equations , because the bernoulli variable and the entropy function are constant along streamlines , the physical variables @xmath19 do not converge to the background state along the streamlines .
they converge only across the streamlines away from the wedge .
therefore , when the elliptic estimates are performed , the scaling is with respect to the distance from the wedge , rather than @xmath15 .
this results in the following decay pattern : in lagrangian coordinates @xmath20 , there exists an asymptotic limit @xmath21 ; @xmath9 converges to @xmath22 at rate @xmath23 , but @xmath24 converges at rate @xmath25 .
that is , the extra decay for the derivatives is only along the @xmath26direction .
finally , we remark that our analysis of transonic shocks for the euler equations for potential and non - potential flow , started in chen - feldman @xcite to formulate the transonic shock problems as one - phase free boundary problems , is motivated by the previous works on variational one - phase free boundary problems for nonlinear elliptic equations in alt - caffarelli @xcite , alt - caffarelli - friedman @xcite , and the references cited therein .
one of the main difficulties in dealing with the transonic shock problems is that the corresponding elliptic one - phase free boundary problems are non - variational in general , so that the complete solution to the free boundary problems requires different approaches and new techniques which are further developed in this paper in the physical realm of the full euler equations for compressible fluids .
the rest of the paper is organized in the following sections . in
[ sec - setup ] , the wedge problem is formulated as a free boundary problem and the main theorem is stated . in [ sec - lagrange ] , the problem is reformulated in lagrangian coordinates . in [ sec - decompose ] ,
the euler equations are decomposed into two algebraic equations and a first - order elliptic system of two equations . in
[ sec - linearize ] , the linear elliptic system and the boundary conditions for iterations are introduced . in
[ sec - keylemma ] , the key estimates of solutions for the linear second - order elliptic equation for iterations are obtained . in [ sec - iteration ] ,
the iteration map is constructed and the corresponding estimates are obtained , leading to the existence of a weak transonic shock solution . in [ sec - unique ] , the uniqueness of the weaker transonic shock solution is proved . in [ sec - asymptotic ] , the asymptotic behavior and the decay rate of solutions are discussed . in [ sec - th ] , the difference between the weaker and the stronger transonic shocks is revealed in terms of the estimates and the asymptotic behavior of the solution .
in this section , we formulate the transonic wedge problem as a free boundary problem and state the main theorem .
the governing equations are two - dimensional steady , full euler equations : @xmath27 where @xmath28 is the gradient in @xmath29 , @xmath30 the velocity , @xmath31 the density , @xmath32 the pressure , and @xmath33 the adiabatic exponent , as well as @xmath34 is the energy .
the sonic speed of the flow is @xmath35 the flow is subsonic if @xmath36 and supersonic if @xmath37 . for a transonic flow ,
both cases occur in the flow .
system can be written in the following general form as a system of conservation laws : @xmath38 with @xmath39 .
such systems often govern time - independent solutions for multidimensional quasilinear hyperbolic systems of conservation laws ; _ cf .
_ dafermos @xcite and lax @xcite . to be a weak solution of the euler equations , the rankine - huguoniot conditions must be satisfied along the shock - front @xmath40 : @xmath41={\sigma}'(x_2 ) [ \,\rho u_2\,],\\ & [ \,\rho u_1 ^ 2 + p\,]= { \sigma}'(x_2 ) [ \,\rho u_1 u_2\,],\\ & [ \,\rho u_1 u_2\ , ] = { \sigma}'(x_2 ) [ \,\rho { u_2}^2 + p\,],\\ & [ \,\rho u_1(e+\frac{p}{\rho})\ , ] = { \sigma}'(x_2 ) [ \,\rho
u_2 ( e+\frac{p}{\rho})\ , ] , \end{aligned } \right.\ ] ] where @xmath42 $ ] denotes the jump of the quantity between the two states across the shock front ; that is , if @xmath43 and @xmath44 represent the left and right states , respectively , then @xmath45:= w^+ - w^-$ ] . for
a given constant upstream supersonic flow @xmath46 and a fixed straight - sided wedge with wedge angle @xmath47 , the downstream constant flow can be determined by the rankine - huguoniot conditions . according to the shock polar ( see fig .
[ fig - shockpolaruv ] ) , there are two subsonic solutions ( for a large - angle wedge ) , or one subsonic solution and one supersonic solution ( for small - angle wedge ) .
we choose the subsonic constant state for the downstream flows .
when the wedge angle @xmath47 is between @xmath48 and the detachment angle @xmath49 , arc @xmath50 is divided by the tangent point @xmath51 into two open arcs @xmath1 and @xmath0 , which correspond to the strong and weak transonic shocks , respectively . for convenience ,
we rotate the plane by angle @xmath47 so that the downstream flows become horizontal . then @xmath52 , @xmath53 , and @xmath54 ( _ cf . _
[ fig - perturbedshock ] ) .
suppose that the background shock is the straight line given by @xmath55 .
let @xmath56 be the region for the upstream flows defined by @xmath57 we use a function @xmath58 to describe the wedge boundary : @xmath59 along the solid wedge boundary @xmath60 , the slip condition is satisfied : @xmath61 suppose that the shock front @xmath62 we seek is @xmath63 then the domain for the subsonic flow is denoted by @xmath64 and @xmath65 in eulerian coordinates , height=219 ] therefore , the problem can be formulated as the following free boundary problem : let @xmath66 be a constant transonic solution with transonic shock @xmath67 . for any upstream flow @xmath68 for equations in domain @xmath69 as a small perturbation of @xmath70 , find a subsonic solution @xmath9 and a shock - front @xmath71 , which are close to @xmath72 and @xmath67 , respectively , such that 1 .
@xmath9 satisfies equations in domain @xmath73 ; 2 .
the slip condition holds along the boundary @xmath60 ; 3 .
the rankine - hugoniot conditions as free boundary conditions hold along the shock - front @xmath71 .
when @xmath72 corresponding to a state on arc @xmath0 gives a weak transonic shock , the problem is denoted by * problem wt * , while the strong transonic shock problem corresponds to arc @xmath1 , denoted by * problem st*. to state our results , we need to introduce the weighed hlder norms for our subsonic domain @xmath74 , where @xmath74 is either a truncated triangular domain or an unbounded domain with the vertex at origin @xmath75 and one side as the wedge boundary .
there are two weights : one is the distance function to origin @xmath75 , and the other is to the wedge boundary @xmath76 . for any @xmath77 ,
define @xmath78 & { \delta}^{\rm w}_{\mathbf{x}}:= \min(\textrm{dist}({\mathbf{x}},\partial \mathcal{w}),1 ) , & { \delta}^{\rm w}_{{\mathbf{x}},{\mathbf{x } } ' } : = \min ( { \delta}_{\mathbf{x}},{\delta}_{{\mathbf{x } } ' } ) , \\[1 mm ] & { \delta}_{\mathbf{x}}:= |{\mathbf{x}}|+1 , & { \delta}_{{\mathbf{x } } , { \mathbf{x}}'}:=\min({\delta}_{\mathbf{x } } , { \delta}_{{\mathbf{x}}'}),\\[1 mm ] & \widetilde{{\delta}}_{\mathbf{x}}:= \textrm{dist}({\mathbf{x}},\partial \mathcal{w})+1 , & \widetilde{{\delta}}_{{\mathbf{x } } , { \mathbf{x}}'}:=\min(\widetilde{{\delta}}_{\mathbf{x } } , \widetilde{{\delta}}_{{\mathbf{x}}'}).\end{aligned}\ ] ] let @xmath79 , @xmath80 with @xmath81 , and @xmath82 be a nonnegative integer .
let @xmath83 be an integer - valued vector , where @xmath84 , @xmath85 , and @xmath86 .
we define @xmath87_{k,0;(\tau , l);e}^{({\gamma}_1;{{\mathbf o}})({\gamma}_2;\partial \mathcal{w})}\nonumber\\ & \quad : = \sup_{\begin{subarray}{c } { \mathbf{x}}\in e\\ |{\mathbf{k}}|=k \end{subarray } } \begin{array}{l } \big\{({\delta}^{\rm o}_{{\mathbf{x}}})^{\max\{{\gamma}_1 + \min\{k,-{\gamma}_2\},0\}}({\delta}^{\rm w}_{{\mathbf{x}}})^{\max\{k+{\gamma}_2,0\ } } \,{\delta}_{\mathbf{x}}^{\tau } \widetilde{{\delta}}_{\mathbf{x}}^{l+k } |d^{\mathbf{k}}f({\mathbf{x}})|\big\ } , \end{array}\label{def - normk0}\\[2 mm ] & { [ f ] } _ { k,{\alpha};(\tau , l);e}^{({\gamma}_1;{{\mathbf o}})({\gamma}_2;\partial \mathcal{w } ) } : = \sup _ { \begin{subarray}{c } { \mathbf{x } } , { \mathbf{x}}'\in e\\ { \mathbf{x}}\ne { \mathbf{x}}'\\ \end{subarray } } \bigg\{\begin{array}{l } ( { \delta}^{\rm o}_{{\mathbf{x}},{\mathbf{x}}'})^{\max\{{\gamma}_1 + \min\{k+{\alpha},-{\gamma}_2\},0\}}({\delta}^{\rm w}_{{\mathbf{x}},{\mathbf{x}}'})^{\max\{k+{\alpha}+{\gamma}_2,0\}}\\[1 mm ] \,\,\,\times { \delta}_{{\mathbf{x}},{\mathbf{x}}'}^{\tau } \widetilde{{\delta}}_{{\mathbf{x}},{\mathbf{x}}'}^{l+k+{\alpha}}\frac{|d^{\mathbf{k}}f({\mathbf{x}})-d^{\mathbf{k}}f({\mathbf{x}}')|}{|{\mathbf{x}}-{\mathbf{x}}'|^{\alpha } } \end{array}\bigg\ } , \label{def - normkga } \\ & \|f\|_{k,{\alpha};(\tau , l ) ; e}^{({\gamma}_1;{{\mathbf o}})({\gamma}_2;\partial \mathcal{w})}:= \sum_{i=0}^k { [ f ] } _ { i,0 ; ( \tau , l);e}^{({\gamma}_1;{{\mathbf o}})({\gamma}_2;\partial \mathcal{w } ) } + { [ f ] } _ { k,{\alpha};(\tau , l);e}^{({\gamma}_1;{{\mathbf o}})({\gamma}_2;\partial \mathcal{w})}. \label{def - norm}\end{aligned}\ ] ] for a vector - valued function @xmath88 , we define @xmath89 let @xmath90 the requirement that @xmath81 in the definition above means that the regularity up to the wedge boundary is no worse than the regularity up to the vertex .
when @xmath91 , the @xmath92terms disappear so that @xmath93 in the superscript can be dropped .
if there is no weight @xmath94 in the superscript , the @xmath95terms for the weights should be understood as @xmath96 and @xmath97 in and , respectively .
when no weight appears in the superscripts of the seminorms in , it means that neither @xmath98 nor @xmath99 is present . for a function of one variable defined on @xmath100 , the weighted norm @xmath101
is understood in the same sense as the the definition above with weight to @xmath102 and the decay at infinity . since the variables in @xmath9 are expected to have different regularity , we distinguish these variables by defining @xmath103 and @xmath104 , where @xmath105 .
let @xmath106 and @xmath107 be the corresponding background subsonic states .
[ thm - main ] there are positive constants @xmath108 , and @xmath109 , depending only on the background states @xmath66 , such that there exists a solution @xmath110 for each of * problem wt * and * problem st * , so that the following estimates hold : 1 . for * problem wt
* , @xmath111 provided that @xmath112 where @xmath113 2 . for * problem st * , @xmath114 provided that @xmath115 where @xmath116 the solution @xmath110 is unique within the class such that the left - hand side of for * problem wt * or for * problem st * is less than @xmath117 .
[ rem:2.2a ] the dependence of constants @xmath108 , and @xmath109 in theorem [ thm - main ] is described as follows : @xmath118 and @xmath119 depend on @xmath70 and @xmath72 , but are independent of @xmath120 and @xmath109 ; @xmath120 depends on @xmath121 , and @xmath122 , but is independent of @xmath123 ; and @xmath123 depends on all @xmath124 , and @xmath120 .
the difference in the results of the two problems is that the solution of * problem wt * has less regularity at corner @xmath125 and decays faster with respect to @xmath15 ( or the distance from the wedge boundary ) than the solution of * problem st*. the asymptotic behavior of the subsonic solution can be stated more clearly in lagrangian coordinates .
thus we leave it in the statement of theorem [ thm - lag ] and remark [ rk - asymp ] .
from the first equation in , there exists a unique stream function @xmath126 in region @xmath127 such that @xmath128 with @xmath129 . to simplify the analysis , we employ the following lagrangian coordinate transformation : @xmath130 y_2= \psi(x_1 , x_2 ) , \end{cases}\ ] ] under which the original curved streamlines become straight . in the new
coordinates @xmath131 , we still denote the unknown variables @xmath132 by @xmath133 for notational simplicity .
the euler equations in in lagrangian coordinates become the following equations in divergence form : @xmath134 & \big(u_1 + \frac{p}{\rho u_1}\big)_{y_1 } -\big(\frac{p u_2}{u_1}\big)_{y_2}= 0 , \label{eqn - euler2}\\[1 mm ] & ( u_2)_{y_1 } + p_{y_2}= 0,\label{eqn - euler3 } \\[1 mm ] & \big(\frac{1}{2}|{\mathbf{u}}|^2 + \frac{{\gamma}p}{({\gamma}-1)\rho}\big)_{y_1}=0 .
\label{eqn - euler4}\end{aligned}\ ] ] let @xmath135 be a shock - front .
then , from the equations above , we can derive the rankine - hugoniot conditions along @xmath136 : @xmath137=-\big[\frac{u_2}{u_1 } \big ] { \hat{\sigma}}'(y_2),\\[1 mm ] & \big [ u_1 + \frac{p}{\rho u_1}\big]=-\big[\frac{p u_2}{u_1}\big ] { \hat{\sigma}}'(y_2 ) , \label{con - rh2}\\[1 mm ] & [ \,u_2 \,]= [ \,p\ , ] { \hat{\sigma}}'(y_2 ) , \label{con - rh3}\\[1 mm ] & \big[\frac{1}{2}|{\mathbf{u}}|^2 + \frac{{\gamma}p}{({\gamma}-1)\rho}\big]= 0 .
\label{con - rh4}\end{aligned}\ ] ] the background shock - front now is @xmath138 , where @xmath139 . without loss of generality , we assume that the supersonic solution @xmath68 exists in region @xmath140 defined by @xmath141 let @xmath142 for a given shock function @xmath143 , let @xmath144 theorem [ thm - main ] can be stated in lagrangian coordinates as follows : [ thm - lag ] there exist positive constants @xmath108 , and @xmath109 , depending only on the background states @xmath70 and @xmath72 such that , if the upstream flow @xmath68 for and the wedge function @xmath145 satisfy 1 .
@xmath146 @xmath147 for * problem wt * , 2 .
@xmath148 @xmath147 for * problem st * , then there exist a transonic shock @xmath135 and a subsonic solution @xmath9 of the euler equations satisfying the rankine - hugoniot conditions along @xmath136 and the slip condition @xmath149 , and there exists a limit function @xmath150 and @xmath151 such that @xmath9 satisfies the following estimates : 1 . for * problem wt
* , @xmath152 where @xmath153 2 . for * problem
st * , @xmath154 where @xmath155 moreover , solution @xmath9 is unique in the class such that the left - hand side of estimate for * problem wt * ) or for * problem st * ) is less than @xmath117 .
see also fig .
[ new-2 ] . and @xmath156 in lagrangian coordinates , height=219 ] [ rk - asymp ] in general , the asymptotic limit @xmath157 is not a constant , which indicates that @xmath19 does not converge to the background state @xmath158 as @xmath159 ( along the streamlines ) ; while @xmath19 converges to the background state as @xmath160 ( transversal to the streamlines away from the wedge ) .
such an asymptotic behavior is owing to the conservation of the bernoulli quantity and the entropy function along the streamlines , which is different from that for potential flow .
estimates in theorem [ thm - lag ] , together with the rankine - hugoniot conditions , imply that the coordinate transformation is bi - lipschitz across the shock - front and has the corresponding regularity in each supersonic or subsonic domain .
therefore , theorem [ thm - lag ] implies theorem [ thm - main ] .
we now use the left eigenvectors to decompose the euler equations into an elliptic system and two algebraic equations .
rewrite system into the following nondivergence form for @xmath39 : @xmath161 where @xmath162 1 - \frac { p } { \rho u_1 ^ 2 } & 0 & \frac{1}{\rho u_1 } & - \frac { p } { \rho^2 u_1}\\[1.5 mm ] 0 & 1 & 0 & 0\\[1 mm ] u_1 & u_2 & \frac{{\gamma}}{({\gamma}- 1 ) \rho } & - \frac{{\gamma}p}{({\gamma}- 1 ) \rho^2 } \end{bmatrix},\\[1 mm ] b(u ) & = \begin{bmatrix } \frac{u_2 } { u_1 ^ 2 } & -\frac{1}{u_1 } & 0&0\\[1.5 mm ] \frac { pu_2 } { u_1 ^ 2 } & - \frac{p}{u_1 } & - \frac{u_2 } { u_1 } & 0 \\[1.5 mm ] 0 & 0 & 1 & 0\\[1 mm ] 0 & 0 & 0 & 0 \end{bmatrix}. \end{aligned}\ ] ] solving @xmath163 for @xmath164 , we obtain four eigenvalues : @xmath165 where @xmath166 in the subsonic region .
the corresponding left - eigenvectors are @xmath167 then 1 .
multiplying equations from the left by @xmath168 leads to the same equation .
this , together with the rankine - hugoniot condition , implies the bernoulli law : @xmath169 in both supersonic and subsonic domains and across the shock - front .
therefore , @xmath170 can be computed from the upstream flow @xmath68 . if @xmath171 is a small perturbation of @xmath172 , then @xmath173 .
therefore , we can solve for @xmath171 : @xmath174 where @xmath175 .
2 . multiplying system from the left by @xmath176 gives @xmath177 3 .
multiplying equations from the left by @xmath178 and separating the real and imaginary parts of the equation lead to the elliptic system : @xmath179 where @xmath180 and @xmath181 therefore , equations
are decomposed into
. we will follow the steps below to solve this problem : \1 .
given a shock - front @xmath182 , introduce a linear system for iterations ; \2 . for a given @xmath9 ,
find @xmath183 by solving the linear system with equations and the corresponding boundary conditions ; \3 . use solution @xmath183 to update the shock - front and obtain @xmath184 , so that we construct a map @xmath14 from @xmath185 to @xmath186 ; \4 .
prove the existence of the solution as a fixed point of @xmath14 by applying the schauder fixed point theorem .
for a given shock - front @xmath182 , the subsonic domain @xmath156 depends on @xmath182 . for the convenience of solving the problem
, we make the following coordinate transformation to change the domain from @xmath156 to @xmath187 : @xmath188 where @xmath189 . in the @xmath190coordinates
, @xmath133 becomes @xmath191 , depending on @xmath182 .
when there is no ambiguity , we may omit the subscript and still denote @xmath191 by @xmath192 .
however , the upstream flow @xmath68 involves an unknown variable explicitly depending on @xmath182 : @xmath193 where @xmath68 is the given upstream flow in the @xmath20coordinates .
hence , equations become the following equations in the @xmath190-coordinates : @xmath194 where @xmath195 using system to solve for @xmath196 yields the linear system for iterations : @xmath197 in the @xmath190-coordinates , the rankine - hugoniot conditions keep the same form , except that @xmath198 is replaced by @xmath199 and @xmath68 is replaced by @xmath200 along line @xmath201 . among the four rankine - hugoniot conditions ,
is used in the bernoulli law . from condition
, we have @xmath202}{[p]}(k_1 z_2 , z_2),\ ] ] which will be used to update the shock - front later .
now , because of , we can use @xmath203 as the unknown variables along @xmath201 . using to eliminate @xmath204 in conditions gives @xmath205 \big [ \frac{1}{\rho u_1 } \big ] + [ w ] [ u_1 w ] = 0 , \label{con - g1}\\ & & g_2(u_{{\hat{\sigma}}}^- , \bar{u } ) : = [ p ] \big [ u_1 + \frac{p}{\rho u_1 } \big ] + [ pw ] [ u_1 w ] = 0.\label{con - g2}\end{aligned}\ ] ] we linearize the conditions above as @xmath206 denoted by @xmath207 where @xmath208 using the two conditions for @xmath209 to eliminate @xmath210 leads to @xmath211 a direct calculation shows @xmath212 \big ( \frac{{\gamma}p_0^+}{({\gamma}-1)(\rho_0^+)^2 u_{10}^+ } + \frac{p_0 ^ -}{u_{10}^-}\big(\frac{1}{(\rho_0^+)^2 } + \frac{{\gamma}p_0^+}{({\gamma}-1)(\rho_0^+)^3 ( u_{10}^+)^2}\big)\big)\\ & & > 0.\end{aligned}\ ] ] therefore , condition becomes @xmath213 where @xmath214 variables , height=245 ] [ signb1 ] the shock polar is a one - parameter curve determined by the rankine - hugoniot conditions .
if @xmath32 is used as the parameter , by equation , we obtain that @xmath215 , which shows that @xmath216 is the linear term and @xmath217 is the higher order term . from fig .
[ wppolar ] , we know that @xmath218 is decreasing in @xmath32 on arc @xmath1 and increasing on @xmath0 . therefore , it is easy to see that @xmath219 corresponds to the state on arc @xmath1 , @xmath220 to @xmath0 , and @xmath221 at the tangent point @xmath51 .
we compute @xmath222\big(-\frac{p_0^+}{(\rho_0^+)^2 u_{10}^+}-\frac{{\gamma}p_0^+}{({\gamma}-1)(\rho_0^+)^3 ( u_{10}^+)^3}\big ) < 0.\ ] ] thus condition for @xmath223 can be rewritten as @xmath224 where @xmath225 , and @xmath226 .
we notice that conditions are equivalent to conditions for @xmath209 .
consider the elliptic equation @xmath227 with boundary conditions : @xmath228 where @xmath187 is the unbounded triangular domain with two boundaries @xmath229 and @xmath201 defined by , and @xmath230 is a constant vector with @xmath231 .
let @xmath232 be the angle between @xmath229 and @xmath201 , and let @xmath233 and @xmath234 be the normal and tangent components of @xmath235 , respectively .
note that @xmath236 is the outer normal to @xmath237 , and @xmath238 is tangent to @xmath237 directed away from the corner on domain @xmath187 .
we assume that @xmath239 .
[ thm - lemma ] consider the boundary value problem . 1 .
when @xmath240 , there exist suitably small @xmath241 , depending only on @xmath235 and @xmath242 , such that , if @xmath243 for a suitably small constant @xmath244 depending only on @xmath245 , and @xmath246 , and @xmath247 @xmath248 , and @xmath249 , then there exists a unique solution @xmath250 of problem . furthermore , there exists a constant @xmath251 , depending only on @xmath245 , and @xmath246 , such that the following estimate holds : @xmath252 2 . when @xmath253 , there exist suitably small @xmath241 , depending only on @xmath235 and @xmath254 , such that , if @xmath255 for a suitably small constant @xmath244 depending only on @xmath245 , and @xmath246 , @xmath256 , and @xmath257 , then there exists a unique solution @xmath258 satisfying the following estimate : @xmath259 where @xmath251 is a constant , depending only on @xmath245 , and @xmath246 . in the following estimates , all constants @xmath260 , etc .
are generic positive constants depending only on the background states @xmath70 and @xmath72 ( or @xmath235 and @xmath261 in lemma [ thm - lemma ] ) , @xmath118 , and @xmath246 .
we first prove part @xmath262 of lemma [ thm - lemma ] .
we truncate domain @xmath187 by line @xmath263 , @xmath264 , into a triangle @xmath265 and prescribe the following boundary condition : @xmath266 since @xmath267 is a bounded domain , we can start with a neumann condition on @xmath201 and dirichlet conditions on @xmath229 and @xmath268 , and then use the continuity method to prove that there exists a unique solution @xmath269 ( _ cf_. theorem 1 in @xcite ) .
the process is standard , based on the _ apriori _ estimates for @xmath270 .
we will focus on obtaining the desired estimates of @xmath270 , independent of @xmath271 .
denote @xmath272 the @xmath3estimates consist of two parts corner estimates and decay estimates .
* corner estimates*. let @xmath273 .
assume @xmath274 ( otherwise , the maximum principle applied to the zero boundary conditions implies a trivial solution ) , and set @xmath275 .
define a comparison function : @xmath276 where @xmath277 are the polar coordinates .
choose @xmath278 suitably small , so that @xmath279 .
now we estimate @xmath280 in the following steps .
first , @xmath281 condition implies that @xmath282 this gives rise to the following estimate : @xmath283 the estimate above yields @xmath284 if @xmath95 is chosen sufficiently small . on the boundaries , we compute @xmath285 & \ge c c_2mr^{{\alpha}-1 } \qquad\,\ , ( \text{by choosing suitably small } { \alpha})\\ & >
g_6 \qquad\,\ , ( \text{by choosing suitably large } c ) , \\[2 mm ] v_1 |_{\mathcal{l}_1 } & = cm z_1^{\alpha}\ge g_5(z_1 ) -g_5(0),\\[1 mm ] v_1 |_{l_r } & \ge g_5(r ) -g_5(0).\end{aligned}\ ] ] therefore , by the comparison principle , we conclude @xmath286 by adding a negative sign to @xmath287 , we obtain that @xmath288 .
thus , we have @xmath289 , we have @xmath290
* decay estimates*. now we estimate the decay rate of @xmath270 in @xmath291 .
denote @xmath292 , and let @xmath293 for @xmath294 , we calculate @xmath295 by adjusting @xmath296 suitably large .
then @xmath297 implies that @xmath298 for sufficiently small @xmath95 .
moreover , using @xmath299 , @xmath300 , and @xmath301 , we have @xmath302 & \ge cmr^{-2-{\mathbf{\beta } } } \qquad\quad ( \text{for large $ c_3 $ and $ c_4$})\\ & > g_6\end{aligned}\ ] ] and @xmath303 by the comparison principle , we conclude latexmath:[\[\label{est - c0decay } which yields the following @xmath3estimate : @xmath305 since we will let @xmath271 approach to @xmath307 eventually , the estimates in @xmath308 will be sufficient .
our estimates are based on the standard schauder interior or boundary estimates in the discs with appropriate scalings ; _ cf . _ gilbarg - trudinger @xcite .
on the other hand , we have different scalings for the corner and away from the corner .
first , we focus on the estimates near corner @xmath75 . for any point @xmath309 with polar
coordinates @xmath310 , we divide the situation into three cases : @xmath311 , @xmath312 , and @xmath313 .
_ case 1 _ : @xmath311 .
let @xmath314 and @xmath315 for @xmath316 .
we rescale @xmath317 into @xmath318 by the coordinate transformation : @xmath319 let @xmath320 and @xmath321 . by the @xmath322estimate near the corner , we have @xmath323 equation
becomes @xmath324 where @xmath325 . since @xmath326 , it is easy to see that @xmath327_{0,{\alpha } ; b_2}= \bar{r}^{{\alpha } } [ a_{ij}]_{0,{\alpha } ,
b_{2\bar{r } } } \le c{\delta},\\[1 mm ] & \|\tilde{a}_{ij}\|_{0,{\alpha } ; b_2 } \le \lambda , \\[1 mm ] & \tilde{a}_{ij}({\mathbf{y}})\xi_i\xi_j \ge { \lambda}|\xi|^2 \qquad \mbox{for $ { \mathbf{y}}\in b_2$}\end{aligned}\ ] ] for suitably small @xmath95 , where @xmath328 are constants depending only on @xmath329 , and @xmath246 .
we apply the schauder interior estimate ( _ cf .
_ theorem 8.32 in @xcite ) to obtain @xmath330 let @xmath331 be a domain and @xmath332 a function defined in @xmath331 , and set @xmath333 .
we define the following norm @xmath334 : @xmath335_{j,0;\omega},\\ & & \| u\|'_{k,{\alpha};\omega } = \| u\|'_{k ; \omega}+ d^{k+{\alpha } } [ u]_{k,{\alpha};\omega}.\end{aligned}\ ] ] then we obtain the estimate for @xmath336 : @xmath337 which implies @xmath338 : @xmath312 .
let @xmath339 , and @xmath340 .
we use the same scaling as in _ case 1_. then the boundary estimates for the poisson equation with the oblique derivative conditions ( see theorem 6.26 in @xcite ) , followed by the technique of freezing the coefficients ( _ cf .
_ lemma 6.29 in @xcite ) , imply that @xmath341 : @xmath313 . now @xmath342 and
@xmath343 are defined in the same fashion as in _ case 2 _ , while @xmath344 .
the schauder boundary estimates for the dirichlet conditions give rise to the @xmath306estimates near boundary @xmath229 ( _ cf .
_ corollary 8.36 in @xcite ) : @xmath345 therefore , estimates in the cases above give the desired corner estimate : @xmath346 now we consider the domain away from the corner : @xmath347 .
the estimates below follow the similar way to the corner estimates , but with a different scaling .
for any @xmath348 , set @xmath349 . then we consider two cases : @xmath350 and @xmath351 . :
@xmath350 . set @xmath352 , @xmath353 , and @xmath354 .
similarly , the schauder boundary estimate yields @xmath355 by using . _ case 2 _ : @xmath351 . set @xmath356 , @xmath357 , and @xmath358 . similar to the @xmath322estimates away from the corner in
[ c0estimates ] , we rescale to the unit disc by the coordinate transformation @xmath359 and then do either the schauder boundary or the interior estimates for @xmath320 . since @xmath360_{0,{\alpha};b_2^+ } = \bar{r}^{{\alpha } } [ a_{ij}]_{0,{\alpha } ; b^+_{2\bar{r } } } \le \bar{r}^{-{\mathbf{\beta}}-1 } [ a_{ij}]_{0,{\alpha};(0,1+{\mathbf{\beta } } ) ; b^+_{2\bar{r } } }
\le c{\delta } , \end{aligned}\ ] ] we obtain the following estimate : @xmath361 where @xmath362 and @xmath363 are the rescaled function of @xmath364 and the rescaled boundary of @xmath51 , respectively . scaling back to @xmath365 leads to @xmath366 estimates give rise to the @xmath306estimate in @xmath367 : @xmath368 combining estimates in @xmath369 with estimate in @xmath367 renders the following @xmath306estimate in @xmath308 : @xmath370 for the @xmath371estimates , we rewrite equation into the following nondivergence form : @xmath372 following the same argument as the @xmath306estimates in [ c1estimates ] , we let @xmath373 , and @xmath374 .
for @xmath375 , we follow the same procedure as in the @xmath306estimates to conclude @xmath376 to obtain the estimates for @xmath377 , we set @xmath378 for any @xmath379 , latexmath:[\[\label{est - vbar } when @xmath377 and @xmath381 , the schauder interior estimate , together with and the @xmath306estimate , leads to @xmath382 finally , for @xmath383 , by the corner estimate , @xmath384_{0,{\alpha};b_{3\bar{r } } } \le cm |{\mathbf{z}}^0|^{-1}.\ ] ] therefore , we have @xmath385 estimates and imply @xmath386 taking @xmath387 , we obtain a sequence @xmath388 .
we can choose a proper subsequence @xmath389 such that @xmath390 converges to @xmath391 in @xmath392 for all @xmath393 , where @xmath394 .
therefore , the limit function @xmath391 is a solution with estimate .
suppose that @xmath395 both are the solutions for problem . then @xmath396 is also a solution of with @xmath397 and @xmath364 vanishing in and , respectively .
@xmath398 implies that @xmath399 decays as @xmath400 . for any small @xmath401
, there exists @xmath402 such that @xmath403 on @xmath268 .
thus , by applying the maximum principle , we see that @xmath404 . we know
that @xmath405 in @xmath187 as @xmath406 .
the procedure of proving part ( ii ) is primarily parallel to that of part ( i ) , except for the different regularity at the corner and decay rate due to the opposite sign of @xmath407 . for the @xmath306regularity at corner @xmath75
, we can estimate @xmath408 , where @xmath409 is solved from the equations : @xmath410 \mathbf{a } \cdot \boldsymbol{\nu } = g_6(0 ) .
\end{cases}\ ] ] since @xmath239 and @xmath411 imply that @xmath412 , the equations above are uniquely solvable for @xmath413 .
once we prove that @xmath270 is @xmath306 up to corner @xmath75 , we can see that @xmath414 .
we use @xmath415 to control @xmath416 near corner @xmath75 .
in fact , denoting @xmath417 , we have @xmath418 by choosing @xmath419 large enough
. then we compute @xmath420 & |\big(a_{ij}\big)_{z_j } ( v_3)_{z_i}| = o({\delta})m ( z_2 + 1)^{-{\mathbf{\beta}}-1}r^{{\alpha } } ( \sin \theta)^{{\alpha}-1 } \le c m{\delta}r^{{\alpha}-1 } ( \sin { \theta})^{{\alpha}-2 } .\end{aligned}\ ] ] the estimates above yield @xmath421 for sufficiently small @xmath95 . on the boundaries , we use @xmath299 , @xmath253 , and @xmath301 to obtain @xmath422 & \ge c_5 c_4 mr^{{\alpha } } \\ & \ge g_6 -g_6(0 )
\qquad\quad ( \text{by choosing suitably large $ c_5$}),\\[2 mm ] v_3|_{\mathcal{l}_1 } & = c_5 m z_1^{1+{\alpha } } \sin { \theta}_0 \ge g_5(z_1 ) -g_5(0 ) -g_5'(0)z_1,\\[1 mm ] v_3|_{l_r } & \ge g_5(r ) -g_5(0)-g_5'(0)r.\end{aligned}\ ] ] thus , by the comparison principle , we conclude @xmath423 on the other hand , the fact that @xmath424 results in the decay rate @xmath425 , which is slower than part ( i ) ( @xmath426 ) .
this can be achieved by setting @xmath427 in the same way as in part ( i ) , we can prove that @xmath428 is a supersolution of .
what is different from part ( i ) is that , for @xmath253 , we require @xmath246 small to guarantee the positivity of @xmath429 on @xmath201 . in fact
, we have @xmath430 which is greater than @xmath364 if @xmath246 is small and @xmath431 is large .
after we obtain the @xmath3estimate , we apply the standard schauder estimates with proper scalings to achieve estimate in part ( ii ) .
we first focus on * problem wt*. for a given upstream flow @xmath68 and @xmath432 in the slip condition satisfying @xmath433 we define a map @xmath14 from @xmath434 to itself , provided @xmath120 and @xmath123 chosen properly , where @xmath434 is given as follows : @xmath435 \sigma_2^\tau & : = \ { v : \| v\|_{2,{\alpha};(1+{\mathbf{\beta}},0);{\mathbb{d}}}^{(-{\alpha};{{\mathbf o}})(-1-{\alpha};\mathcal{l}_1 ) } \le \tau \},\\[1 mm ] \sigma_3^\tau & : = \ { v : \| v\|_{2,{\alpha};(1+{\mathbf{\beta}});{\mathbb{r}}^+}^{(-{\alpha};0 ) } \le \tau \},\\[1 mm ] \sigma^\tau & : = \sigma_1^\tau \times\sigma_1^\tau\times\sigma_2^\tau\times \sigma_2^\tau\times\sigma_3^\tau .
\end{split } \end{aligned}\ ] ] for notational convenience , we use @xmath436 to denote the norm for @xmath437 . the norm @xmath438 is understood as the summation of the norms of all the components . given @xmath439 , we first solve equations with the slip condition @xmath440 and the boundary condition on @xmath201 .
once we obtain @xmath441 , we use condition on @xmath201 and equation to solve for @xmath442 .
then , by , we can compute @xmath443 . from equation
, we update the shock function @xmath444 .
thus , we can define @xmath445 .
we perform @xmath447 @xmath448 to eliminate @xmath449 and obtain @xmath450 where @xmath451 in order to meet condition in lemma [ thm - lemma ] , we apply the following coordinate transformation : @xmath452 \bar{z}_2= \sqrt{\frac{e^0}{{\lambda}^0_i } } z_2 , \end{cases}\ ] ] where @xmath453 are @xmath454 evaluated at the background state @xmath72 .
thus , equation becomes @xmath455 where @xmath456 & \bar{a}_{12}(\bar{{\mathbf{z } } } ) = \bar{a}_{21 } ( \bar{{\mathbf{z}}})= e^0 a_{12}(\sqrt{\frac{1}{e^0 { \lambda}^0_i } } \bar{z}_1 , \sqrt{\frac{{\lambda}^0_i } { e^0}}\bar{z}_2 ) , \quad { \delta}\tilde{\bar{w}}(\bar{{\mathbf{z } } } ) = { \delta}\tilde{w}(\sqrt{\frac{1}{e^0 { \lambda}^0_i } } \bar{z}_1 , \sqrt{\frac{{\lambda}^0_i } { e^0}}\bar{z}_2).\end{aligned}\ ] ] the boundary , @xmath201 , becomes @xmath457 for @xmath458 .
condition becomes @xmath459 in the @xmath460coordinates , where @xmath461 is @xmath462 rescaled in the @xmath460coordinates .
differentiating along @xmath463 and using equations to eliminate the @xmath464 terms gives rise to @xmath465 slightly modify into @xmath466 where @xmath467 \bar{g}_7 & = \bar{g}'_3+\frac{b_1}{e^0}\big((\bar{a}_{11}-1 ) - k_2 \bar{a}_{12}\big ) ( \bar{w})_{\bar{z}_1 } - \frac{b_1}{e^0}\big(k_2(\bar{a}_{22}-1)-\bar{a}_{12}\big)(\bar{w})_{\bar{z}_2}.\end{aligned}\ ] ] conditions and are equivalent when @xmath468 , _ i.e. _ , when @xmath469 is a fixed point of @xmath14 . for * problem wt * , @xmath470 ( see remark [ signb1 ] ) .
then we normalize @xmath471 into @xmath472 and compute @xmath473 moreover , we have @xmath474 for sufficiently small @xmath123 , so that condition is satisfied .
therefore , applying part @xmath262 of lemma [ thm - lemma ] and scaling back to the @xmath190coordinates , we have @xmath475 where @xmath476 is @xmath477 scaled back in the @xmath190coordinates .
we know that @xmath478 since the rankine - hugoniot conditions hold at the background states , we have @xmath479 therefore , @xmath480 defined by can be rewritten as : @xmath481 which gives rise to the following estimates : @xmath482 combining with , estimate becomes @xmath483 choosing @xmath484 and @xmath485 , we have @xmath486 which implies that @xmath487 .
differentiating with respect to @xmath490 yields @xmath491 in domain @xmath492 , we solve the equation : @xmath493 with the following dirichlet boundary conditions : @xmath494 condition is artificially prescribed on @xmath268 so that the continuity of @xmath332 at the intersection points of @xmath268 with @xmath229 and @xmath495 is achieved .
given @xmath271 , we obtain a solution @xmath496 .
the estimates of @xmath496 follow the same way as in lemma [ thm - lemma ] .
once we have the desired _ a priori _ estimates , by the continuity method , we also have the existence of the solution .
therefore , we only need to point out the difference from the _ a priori _ estimates in lemma [ thm - lemma ] .
we use the barrier function @xmath499 where @xmath500 are small so that @xmath501 .
this can be achieved because @xmath502 .
it is easy to see that @xmath503 on the boundary .
following the same computation , we have @xmath504 therefore , we conclude that latexmath:[\ ] ] therefore , we have proved that , given a constant transonic flow on arc @xmath1 or @xmath652 , if the upstream flow and the wedge boundary are perturbed , then there exist a unique subsonic solution and transonic shock , which are close to the background constant state and straight shock front .
this shows the stability of the constant transonic flows past wedges . for the constant states on @xmath652 , the regularity of the subsonic solution near the corner is @xmath653 and the decay rate in the far field is @xmath641 .
furthermore , we gain the higher decay rate @xmath640 for the directional derivative along the streamlines of @xmath654 , the direction of the flow . on @xmath1
, we obtain the @xmath306regularity at the corner and @xmath425 decay in the far field .
* acknowledgments*. the research of gui - qiang g. chen was supported in part by the uk engineering and physical sciences research council award ep / l015811/1 and the royal society wolfson research merit award ( uk ) .
the research of jun chen was supported in part by sustech start - up grant y01286205 .
the research of mikhail feldman was supported in part by the national science foundation under grant dms-1401490 and the van vleck professorship research award by the university of wisconsin - madison .
j. von neumann , discussion on the existence and uniqueness or multiplicity of solutions of the aerodynamical equations [ reprinted from mr0044302 ] , _ bull .
_ , * 47 * ( 2010 ) , 145154 . | the existence , uniqueness , and asymptotic behavior of steady transonic flows past a curved wedge , involving transonic shocks , governed by the two - dimensional full euler equations are established .
the stability of both weak and strong transonic shocks under the perturbation of both the upstream supersonic flow and the wedge boundary is proved .
the problem is formulated as a one - phase free boundary problem , in which the transonic shock is treated as a free boundary .
the full euler equations are decomposed into two algebraic equations and a first - order elliptic system of two equations in lagrangian coordinates . with careful elliptic estimates by using appropriate weighted hlder norms , the iteration map is defined and analyzed , and the existence of its fixed point is established by performing the schauder fixed point argument .
the careful analysis of the asymptotic behavior of the solutions reveals particular characters of the full euler equations . |
self - assembly of block copolymers has been an actively pursued field of study because of its wide technological implications @xcite .
depending upon the volume fraction of the components and segregation regime , block copolymers exhibit a range of periodic morphologies such as lamellae , gyroids , cylinders , spheres etc.@xcite as we intend to study symmetric diblock copolymers which order into lamellar structure , the rest of the paper will focus only on this particular morphology . in the absence of any external field , a symmetric diblock copolymer forms domains of lamellar morphology with various degrees of alignment .
practical applications , however , require complete alignment of the microphase separated domains . in general this can be achieved by application of external fields , primary among them being substrate field @xcite , shear field @xcite and electric field @xcite . since the microphase separation generally takes place on a substrate , it is impractical to assume that the experiments are devoid of surface effects . typically , interfacial energy difference between the two blocks in contact with a substrate ( i.e @xmath0 , where @xmath1 and @xmath2 are the interfacial energies between a monomer a or b and substrate s ) can cause surface induced ordering resulting in parallel arrangement of the domains with respect to the surface @xcite .
if the copolymer system is confined between two rigid substrate walls , two different interaction cases may be considered , ( a ) the walls are symmetric i.e. both walls attract the same monomer and ( b ) antisymmetric walls i.e. both walls attract different monomers @xcite . as shown in fig .
[ fig1 ] ( a ) and ( b ) , the system either forms integral @xmath3 or half - integral @xmath4 number of lamellae , where @xmath5 is the equilibrium lamellar spacing , depending on whether the wall is symmetric or antisymmetric respectively @xcite . if the film thickness is incommensurate with the lamellar spacing , the copolymers are said to be in a frustrated state @xcite .
frustration can be accomodated by deviation of equilibrium spacing from @xmath5 to @xmath6 which is typical in thicker films where the deviation can span over a large number of lamellae @xcite . on the contrary ,
frustration in thin films is relieved by a change of configuration from parallel to perpendicular lamellae @xcite as shown in fig . [ fig1](c ) . in the present work ,
we restrict our study to the case of symmetric walls . electric field , on the other hand is gaining popularity in guiding self - assembly of block copolymers because of the ease with which it can be applied , especially in thin films @xcite .
in contrast to stable parallel configuration in presence of substrate interaction , presence of electric field makes the perpendicular arrangement more stable .
the reason for this can be rationalised in terms of dielectric permittivity mismatch between the two monomer components @xcite . in electrostatics ,
the electric field isolines prefer to pass through the regions of higher permittivity @xcite .
since this is readily achieved in a perpendicular state , the system minimizes its free energy by adopting perpendicular configuration . a schematic showing the various stable configurations in a symmetric diblock copolymer in presence of parallel walls / substrate ( denoted by hatched lines ) .
( a ) parallel lamellae in presence of symmetric wall interaction , ( b ) parallel lamellae in presence of antisymmetric wall interaction , ( c ) perpendicular lamellae arising either due to frustration or applied electric field . ]
a lot of factors can aid or deter the evolving phase morphology in presence of electric field such as the segmental interaction or segregation @xmath7 ( @xmath8 is the flory interaction parameter which scales inversely with temperature and @xmath9 is the degree of polymerization ) , the substrate interaction strength and the confinement which can act as geometric barrier .
the copolymers can be easily modulated at low segregation owing to weak segmental interaction between the two blocks @xcite .
the substrate interaction can hinder the alignment process by electric field as then the applied field has to overcome the interfacial interaction @xcite .
confinement ( film thickness ) on the other hand can aid or hinder the alignment process by electric field depending upon its relation with the natural lamella period @xmath5 @xcite . the existing literature on parallel to perpendicular lamella transition
can be grouped into three categories depending upon the methodology of study : analytical @xcite , experimental @xcite and numerical @xcite . while most of the earlier theoretical work in this regard focus on strong segregation limit ( ssl ) , @xcite literature on weak segregation limit ( wsl ) @xcite is relatively sparse .
a common finding in the ssl is the presence of three stable configurations , namely , parallel , perpendicular and mixed with respect to the substrate .
the mixed structure contains parallel lamellae closer to the substrate and perpendicular lamellae in the middle , depending upon substrate interaction strength and electric field @xcite . if @xmath10 , the electric field causes a direct transition from parallel to perpendicular state @xcite .
however , if , @xmath11 , two critical fields establish in thicker films @xcite . at lower critical field ,
the parallel arrangement changes to mixed morphology , while , on increasing electric field results in perpendicular lamellae @xcite . in wsl ,
the mixed morphology is not predicted and the transition from parallel to perpendicular is direct @xcite .
experimental studies @xcite focusing on intermediate to strong segregation regimes have reported some observations , which do not completely converge with analytical findings .
it is reported that a completely perpendicular arrangement in presence of electric field can only occur when the interfacial energy difference between the two blocks with the substrate is balanced @xcite i.e. @xmath12 .
any mismatch results in a mixed morphology @xcite .
the discrepancy between the analytical and experimental observations are attributed to the pathway dependence of the alignment process in the presence of an electric field @xcite .
analytical studies on the other hand focus only on the final equilibrium configuration . in reality , however , the microdomains can get kinetically trapped in metastable states and a high activation energy may be needed to reach the stable state . thus , numerical studies can provide an efficient bridge between the analytical calculation and experiments .
dynamic density functional theory ( ddft ) @xcite , self - consistent - field theory ( scft ) @xcite and cell dynamics simulations ( cds ) @xcite have previously been employed to study electric field induced alignment .
the focus of most numerical studies have either been on mechanism of alignment of ordered domains @xcite or on order - order transition from one morphology to the other in bulk samples @xcite .
the effect of substrate interaction and confinement have not been investigated thoroughly which could significantly alter the phase morphologies and critical electric field for transition .
matsen studied the stability of monolayer @xcite and multilayer @xcite lamellae films but did not consider substrate affinity .
lyakhova et al .
@xcite and kyrylyuk et al .
@xcite did consider substrate interaction , but the effect of film thickness was not considered .
moreover , the presence of any intermediate or mixed phases were not observed .
thus , the primary objective of the present work is to systematically investigate the effect of substrate interaction and confinement on the resulting morphologies .
we adopt a coarse - grained cahn - hilliard approach to study the self - assembly of symmetric diblock copolymers under competing substrate interaction , electric field and confinement . secondly , one of the serious shortcoming of most previous works involve an approximate solution of the electrostatic field .
the perturbed solution of the maxwell equation is based on the assumption of weak fractional variation of the dielectric constant @xcite . though appropriate in the proximity of the order - disorder transition(odt ) temperature
, the results can be significantly marred as the segregation increases .
such an assumption is relaxed in the present study by coupling the ohta - kawasaki functional to maxwell equation to calculate the electrostatic field distribution .
the organization of the paper is as following . in the next section we present our diffuse interface model , followed by a presentation of our results in section [ sec3 ] .
we conclude the paper by comparing the results to that from experiments and scft calculations in section [ sec4 ] .
the diffuse interface approach to model block copolymers follows the ohta - kawasaki free energy functional @xcite , which includes a long - range interaction term in addition to short range interaction terms ( bulk + interfacial energy ) in the cahn - hilliard model @xcite to account for chain interactions .
the bulk part of the free energy functional of a diblock copolymer can be written as @xcite , @xmath13 d\textbf{r } \\
\nonumber & + b\int_{v } \int_{v } g(\textbf{r},\textbf{r}^\textbf{\ensuremath{\prime}})\psi(\textbf{r})\psi(\textbf{r}^\textbf{\ensuremath{\prime}})d\textbf{r } d\textbf{r}^\textbf{\ensuremath{\prime}}\end{aligned}\ ] ] where the terms in the first integral constitute the short - range interactions , while the second integral consists of the long - range interaction .
@xmath14 is an order parameter that denotes the local concentration difference between the two components , @xmath15 .
@xmath16 is the bulk free energy taken to be of the form @xmath17 to account for two stable phases ( @xmath18 ) below the critical temperature .
@xmath19 is the gradient energy coefficient which penalises gradients in the order parameter .
@xmath20 is a numerical parameter that determines the extent of microphase separation and scales as @xmath21 , where n is the number of segments .
g is the green s function having the property @xmath22 = @xmath23 .
the parameters @xmath19 and @xmath20 are related to the polymer architecture through the relations @xcite , @xmath24 where @xmath25 is the kuhn statistical length or the average monomer space size and @xmath26 is the relative molecular weight which is a measure of the length of a monomer chain compared to the whole macromolecule .
the segregation @xmath7 is determined by @xmath19 and @xmath20 as @xcite , @xmath27 the free energy of the domain surface in presence of attracting walls is written as , @xcite @xmath28 d\textbf{r } .\end{aligned}\ ] ] the above expression results from a taylor series expansion of bare surface energy @xcite .
the terms @xmath29 and @xmath30 have special physical interpretation .
@xmath29 denotes the surface chemical potential difference .
a positive value expresses preferential attraction of b component and vice - versa @xcite .
the term @xmath30 takes into account the deviation of flory parameter @xmath8 at the surface @xcite to account for two confining walls at two ends in the current study , we rewrite the above expression in terms of a @xmath31-function as , @xmath32\psi(\textbf{r})d\textbf{r } , \end{aligned}\ ] ] where , @xmath33 and @xmath34 are the interaction strengths of the wall at @xmath35 and @xmath6 respectively . in the present study the term @xmath30 is set to zero , i.e. we neglect any deviation of interaction from the bulk .
this specific choice of surface potential results in short range interaction .
the electrostatic contribution to the free energy functional can be obtained as @xcite , @xmath36 where , @xmath37 is the permittivity of free space , @xmath38 is the volume occupied by one polymer chain and @xmath39 is the dielectric permittivity which is taken to be phase dependent .
@xmath40 is the space dependent potential due to the applied voltage .
we apply a linear interpolation of the permittivity between the two phases assuming the polymer to behave as a linear dielectric material by , @xmath41 the assumption of linear dielectric behaviour has been previously employed in scft calculations @xcite .
thus the total free energy functional in units of @xmath42 can be finally written as , @xmath43 substituting the derived expressions we get , @xmath44 + \frac{\kappa}{2}|\nabla \psi| ^{2 } + \left [ h_{o}\delta(y ) + h_{l}\delta(y - l)\right]\psi(\textbf{r } ) -
\frac{\epsilon_ov_o^3}{k_bt}\frac{\epsilon(\psi)}{2}|\nabla \phi| ^2 d\textbf{r } \\
\nonumber + b\int \int g(\textbf{r},\textbf{r}^\textbf{\ensuremath{\prime}})\psi(\textbf{r})\psi(\textbf{r}^\textbf{\ensuremath{\prime}})d\textbf{r } d\textbf{r}^\textbf{\ensuremath{\prime } } .\end{aligned}\ ] ] in absence of the electrostatic free energy contribution , the model reduces to the one from brown et al .
@xcite , that was used to study surface induced ordering of block copolymers . as evident , the free energy functional is dependent upon @xmath45 , @xmath40 and their spatial derivatives .
the minimization of the functional , then , requires the evaluation of the variation with respect to each individual variable i.e. @xmath46 and @xmath47 @xcite .
the kinetic evolution of the conserved order parameter @xmath45 follows the dynamics of model b framework @xcite , @xmath48 where the chemical potential @xmath49 is defined as , @xmath50 hence the kinetic equation can be expressed as , @xmath51 - b\psi .\end{aligned}\ ] ] additionally , the variation of the functional with respect to @xmath40 , assuming the electric field to be in local equilibrium , leads to @xcite , @xmath52 = 0\end{aligned}\ ] ] the above expression is nothing but the maxwell equation which provides the spatial distribution of @xmath40 .
we remark that thermal fluctuations can be accounted by adding a noise term @xmath53 in eq .
[ 0 ] , where , @xmath54 is the strength of the noise , the reciprocal of which is roughly equal to the quench depth and @xmath55 is noise distribution following the fluctuation - dissipation theorem @xmath56 .
it is worth mentioning that thermal fluctuations are important as far as the order - disorder transitions in weak segregation regime are concerned .
however , the equilibrium morphology which is the focus of the present work is not influenced upon incorporation of stochastic noise .
the present claim is corroborated by ref .
@xcite , where the effect of confining surfaces was studied and the final morphology was found to be independent of noise , both in wsl as well as ssl . therefore , we assert that the neglect of stochastic noise in the present work is reasonably well justified . a schematic showing the simulation set up used in the present study .
a diblock copolymer thin film is confined within two rigid substrates ( top and bottom ) which has a preferential attraction towards one of the components .
electrostatic field is generated by applying a constant voltage across the film thickness . ]
we non - dimensionalize all the quantities selecting characteristic energy scale @xmath57 , length scale @xmath58 and time @xmath59 , the free energies are rescaled using @xmath60 , @xmath58 by the lattice cell size and the time by @xmath61 . the terms @xmath19 and @xmath20 is non - dimensionalized as , @xmath62 and @xmath63 .
the dielectric constants are rendered non - dimensional using @xmath64 and the electric field by @xmath65 . using representative values of @xmath66 and @xmath67
, we obtain @xmath68 . the simulation geometry chosen for the present study
is shown in fig .
the set - up consists of two rigid surfaces at @xmath69 and @xmath70 confining the copolymer film across which a constant voltage is applied .
dirichlet boundary conditions are applied for voltage at @xmath69 and @xmath70 with @xmath71 and @xmath72 while neumann boundary condition is applied at @xmath73 and @xmath74 . therefore , electric field is aligned along y - direction .
the confining substrates , also attract one of the copolymers which is controlled by the numerical parameter @xmath75 as mentioned before . the appropriate boundary condition to account for attracting substrates translate into , @xcite @xmath76
additionally , no mass transport is allowed through the rigid surface by applying a no - flux boundary condition at the surfaces , @xcite @xmath77 periodic boundary condition is applied for @xmath45 in x - direction .
we solve eq .
( [ 1 ] ) using an explicit finite difference method where the spatial derivatives are discretized using central difference which is second order accurate in space and temporal discretization using first order euler technique .
the laplace equation in eq .
( [ 2 ] ) is solved iteratively using successive - over - relaxation ( sor ) method .
the initial guess for @xmath40 is tailored by providing a linear initial profile in y - direction ( corresponding to constant electric field , since , @xmath78 ) to facilitate faster convergence .
the various non - dimensional model parameters are selected as @xmath79 , @xmath80 , @xmath81 , @xmath82 , @xmath83 , @xmath84 , @xmath85 .
we remark , that the present results are not influenced by the choice of grid resolution ( @xmath86 and @xmath87 ) . to this end
, we are able to replicate our numerical results ( at @xmath88 ) with finer grid spacing ( @xmath89 ) . in order to scale up the time - step width @xmath90 which scales as @xmath91 for cahn - hilliard equation
, we conveniently resort to a larger grid spacing .
the values of the permittivity closely resembles a ps - pmma copolymer system @xcite , though other values have also been used in the literature @xcite .
the values of @xmath19 and @xmath20 correspond to a segregation of @xmath92 .
the surface interaction strength @xmath75 is varied as 0.1 , 0.5 , 1.0 and 1.5 .
since we intend to study symmetric walls i.e. both the surfaces attract the same monomer , we select @xmath93 .
moreover , we are interested to study the effect of confinement .
simulations are carried out for different box sizes in y - direction , @xmath94 .
the natural lamellar spacing is around @xmath95 grid points , so that the selected film thicknesses allows us to study systems with @xmath96 , @xmath97 and @xmath98 .
another implication of the above mentioned values of substrate interaction and film thickness is that the surface induced ordering length is greater than the film thickness . in other words
, this implies that in the absence of electric field , lamellae , parallel to the substrate , span across the entire film .
the surface induced ordering length for the smallest substrate affinity of @xmath99 is around @xmath100 .
the magnitude of electric field is tuned by changing the value of applied voltage , and by normalising it with the box size @xmath101 ,
i.e. , @xmath102 to maintain the same electric field for different box sizes . the box size in x - direction is kept fixed as @xmath103 in all the simulations .
the initial microstructure is generated by assigning a computational noise between @xmath104 about the average composition ( @xmath105 ) corresponding to a disordered state .
the system is then allowed to evolve in presence of electrostatic field and attracting substrates . to gain insights during the microstructure evolution process we define two parameters , average density profile along y direction @xmath106 @xcite and degree of alignment @xmath107 @xcite as , @xmath108 where @xmath109 and @xmath110 are the fourier space wave vectors in x and y direction and @xmath111 = @xmath112 .
@xmath113 is the magnitude of intensity of the fourier power spectrum defined as @xmath114 .
@xmath6 denotes the system size @xmath115 and the terms in the angular bracket imply the product of @xmath45 and its complex conjugate in fourier space . in cases , where the alignment is parallel to the substrates i.e parallel lamellae along the y direction form , @xmath116 , as there is no relevant periodicity along this direction . as a result ,
the value of degree of alignment parameter is @xmath117 . in the opposite case ,
the alignment is perpendicular to the surface , @xmath118 and the value of degree of alignment parameter is @xmath119 .
thus , a value of @xmath117 implies @xmath120 parallel lamellae , while , a value of @xmath119 implies a @xmath120 perpendicular lamellae . a parallel lamellae arrangement in y direction
is characterized by an oscillatory average density profile , whereas a flat profile about @xmath121 corresponds to a perpendicular lamellar arrangement .
we first study some typical morphologies arising due to the interplay of substrate interaction , confinement and electric field followed by an evaluation of the resulting phase diagram .
the result is categorized into three different regimes depending on the film thickness @xmath101 .
the combined effect of the substrate and electric field for model parameters @xmath99 , @xmath123 and @xmath124 is presented in fig .
the phase separation initiates from the surface leading to the formation of parallel lamellae .
however , at t = 80 , the effect of electric field sets in , leading to undulations which ultimately break up the inner layers into smaller domains .
subsequently , the smaller domains coalesce and get aligned in the direction of the electric field
. this phenomenon proceeds outwards layer by layer resulting in a perpendicular lamellar arrangement due to energetic consideration . the average density profile in fig .
[ fig3](e ) during early stages ( corresponding to t = 10 ) is oscillatory near the surface , due to the formation of enriched and depleted layers . with time
the oscillatory profile develops throughout the film thickness . at t = 180 ,
the innermost oscillation dies out and is replaced by a flatter profile which highlights the destruction of parallel structure at the center of the film . much later ,
( corresponding to t = 5000 ) the density profile becomes flat in the bulk of the film .
however the average value shows small enrichment layers at the immediate vicinity of the surfaces even though the microstructure at the final timestep appears to be completely perpendicular .
the kinetics of alignment is presented in fig . [ fig3](f ) .
the value of @xmath125 is -1 during early stages corresponding to parallel ordering along the surface .
there is a smooth temporal transition from -1 to a value closer to + 0.9 which depicts the formation of perpendicular lamellae . the influence of increasing the magnitude of substrate interaction strength and electric field for the same film thickness is presented in fig .
the mechanism that leads to the formation of perpendicular lamellae is essentially the same as earlier , i.e local lamellae disruption and coalescence .
however , as a result of greater substrate interaction , the parallel lamellae near the surfaces i.e. the wetting layer never break resulting in alternate enriched and depleted layers at the boundaries . in literature such a morphology is termed as mixed @xcite . though the surface induced ordering length is greater than the film thickness of our study , the effect of surface is predominant closer to the walls and fades as we move away from the walls . in other words
this implies that the effect of surface is non uniform over the whole ordering length . as a result ,
when the electric field drives the domain alignment perpendicular to the surface , above a threshold interaction strength @xmath75 , and below a threshold electric field @xmath126 , the substrate interaction dominates near the walls resulting in a few parallel layers . meanwhile the effect of electric field is predominant at the center ( away from the wall ) and is able to induce a change in configuration in this region .
the degree of alignment achieved in the direction of electric field is around 75@xmath127 in fig . [
fig4](e ) .
due to higher magnitude of electric field , faster kinetics is observed as can be seen by either comparing the microstructures in fig .
[ fig3 ] and fig .
[ fig4 ] or by comparing the slope of @xmath125 in fig .
[ fig3](e ) and fig .
[ fig4](e ) during the transition period .
the latter has a steep transition region as compared to the smoother transition region in the former case .
finally we discuss the consequences of further increasing the electric field to @xmath128 whilst keeping the other two parameters @xmath129 and @xmath130 unaltered .
the results are shown in fig .
an interesting phenomenon to observe is the mechanism of alignment by the electric field .
in contrast to the previous cases , the parallel ordering never goes beyond two layers . instead , electric field is sufficiently high to orient the composition fluctuations in the non - phase separated region , leading to the appearance of perpendicular lamellae at the middle of the film .
subsequently , the parallel layers near the walls also collapse and the rearrangement of perpendicular lamellae proceeds by defect annihilation mechanism @xcite .
the average density profile in fig . [ fig5](e ) shows enrichment layers at the walls at all times but the oscillatory profile , characteristic of the parallel lamellae configuration never develops at the film center .
the transition regime of alignment kinetics is abrupt as compared to the earlier cases .
the value of @xmath125 saturates to a value of + 0.8 which constitutes to 90@xmath127 alignment in the direction of the applied field .
the next study focuses on configurations when the film thickness is comparable to the bulk lamellar spacing .
in particular we discuss the case of film thickness of @xmath132 which is approximately equal to @xmath133 . for @xmath132 two cases with @xmath99 ( fig .
[ fig6](a ) ) and @xmath129 ( fig .
[ fig6](b ) ) are presented .
the electric field is kept constant at @xmath134 . for low substrate interaction strengths ,
a parallel arrangement is found to be stable as in fig . [ fig6](a ) . however , with increasing substrate interaction strength we see , circular domains emerging at the center of the film as in fig . [ fig6](b ) . to differentiate the effect of electric field and substrate interaction responsible for this phenomenon
, we consider the microstructural pattern in the absence of electric field .
the resulting evolution ( not shown ) is similar , comprising of an inner layer of circular domains .
therefore , it can be inferred that the effect is solely driven by the substrate confinement independently of the applied electric field .
the results can also be interpreted in terms of interference of composition waves . for higher @xmath75 ,
the film thickness , @xmath132 , is close to half integral of equilibrium lamellar spacing .
a destructive interference takes place at the center due to the composition waves emanating from the walls and an inner lamellae can not be maintained . as a result circular domains start appearing in the middle .
the breakup of the inner lamellae into circular domains is similar to the formation of holes as described by the mean - field theory of shull @xcite .
the same phenomenon is absent at low @xmath75 , presumably because the destructive interference at the film center is not sufficiently strong .
an average density plot including the transition at @xmath121 for the case of lower @xmath75 is presented in fig.[fig6](d ) .
it can be verified that the innermost layer is thinner ( @xmath135 ) than the next two adjacent layer on both sides(@xmath136 ) and a parallel arrangement can still be maintained with the innermost layer being in a compressed state .
average density profiles at different electric field strengths for @xmath137 .
all the plots correspond to @xmath138 . ]
the average density profile presented in fig .
[ fig6]e ) shows the asymmetry in the roots of @xmath121 .
the average @xmath139 value at the center is slightly b rich , clearly signifying the absence of any parallel lamellae structure .
the slight asymmetry in the average density points indicates , either the evolution of a perpendicular phase or an in plane asymmetric phase .
the microstructure in fig .
[ fig6]b ) indicates the second possibility where circular domains coexist at the center simultaneouly with the wetting layers .
the presence of such hybrid structures ( combination of two different phases ) have previously been reported in cylinder forming systems at similar film thickness @xcite .
however their transition in electric field has not been reported previously .
although a 3d simulation is desirable to adequately address the issue , we can certainly make some predictions from the current 2d study . the average density profile for @xmath137 at different field strengths is presented in fig .
all the plots correspond to @xmath138 . at low electric field strengths , @xmath134 ,
the plot is similar to fig .
[ fig6]d ) , comprising of parallel lamellae . with a slight increase in the electric field strength , the value of @xmath139 at the center of the film shifts towards zero ( slightly b rich ) .
the density profile is similar to fig .
[ fig6]e ) , and denotes the appearance of a structure other than lamellae . in this case it is not the effect of substrate and confinement alone that causes this transition , but the presence of electric field does play its part . with further increase of electric field @xmath140 and @xmath141 , the surface enrichment decreases . at the same time , the average value at the center of the film shifts to positive values and the profile tends to get flatter .
though it is clear that perpendicular phases now span , atleast in the middle of the film , the exact nature is very difficult to determine precisely in 2d simulation .
we speculate that the perpendicular phases are either cylindrical structures ( for @xmath140 ) or perpendicular lamellae ( for @xmath142 ) .
next , we decrease the film thickness below the equilibrium lamellar spacing , to @xmath143 .
the results corresponding to model parameters @xmath99 and @xmath134 are shown in fig .
the early stage microstructure corresponding to t = 10 is a superposition of parallel and perpendicular lamellae . with time
( t= 30 and 70 ) , the system evolves through a metastable antisymmetric configuration and transforms into a perpendicular state .
this observation is also corroborated by the average density profile .
the result is consistent with the findings of walton et al .
@xcite who argue in favor of a transient antisymmetric arrangement in symmetric thin films during the formation of vertical configuration .
even with the smallest electric field strength ( used in the course of this study ) , a perpendicular arrangement is seen to be stable . even in the absence of the electric field ,
a stable perpendicular arrangement establishes implying that the geometrical confinement predominates over the electric field .
interestingly , @xmath125 reaches the value of + 1 in this case .
this observation points towards the following important fact : when the arrangement is guided by substrate confinement , deviation from perfect perpendicular morphology is negligible .
however , significant deviation in perpendicularity is observed when the ordering is achieved due to the application of electric field . for @xmath144
, parallel arrangement is found to be stable in the absence and at low strength of electric fields . at higher electric field strengths ,
the microstructure is a superposition of parallel and perpendicular lamellae as can be seen in fig .
[ fig9 ] . because of the small film thickness , the substrate interaction is predominant and electric field is not able to completely eradicate the previous surface ordering phenomenon .
correspondingly @xmath125 saturates to a value of 0 which is midway between parallel and perpendicular configuration .
we construct a phase diagram to summarize the influence of electric field @xmath126 , surface interaction strength @xmath75 and film thickness @xmath101 on the evolving morphologies by classifying them into parallel , perpendicular and mixed category .
we designate the morphology mixed only when there exists at least one completely parallel layer ( @xmath145 since the layers closer to the substrate are one - half of the inner layers ) .
classification based on such criteria will allow us to compare the resulting phase diagram with the analytical calculations @xcite .
any other combination of phases is denoted as hybrid structure .
the resulting configuration stability diagram is shown in fig .
[ fig10 ] .
the following points can be appreciated , 1 . for a given film thickness ,
the magnitude of applied electric field to induce a perpendicular arrangement increases with increasing magnitude of substrate interaction strength .
any deviation from this generality occurs only for very thin films e.g. @xmath146 ( fig .
[ fig10](e ) ) and closer to half - integral lamellar thickness i.e @xmath147 ( in our case @xmath132 , fig .
[ fig10](c ) ) .
2 . at low strength of substrate interaction ( @xmath148 ) and electric fields ,
parallel arrangement is found to be stable . however beyond a certain critical value of the electric field e.g. @xmath123 corresponding to @xmath99 , the configuration oscillates between perpendicular and parallel configuration ( fig .
[ fig10 ] ( a)-(e ) ) .
the present findings accentuate the previous analytical results @xcite . in general : as @xmath101 decreases , the effect of substrate interaction becomes more prominent .
therefore a higher magnitude of electric field is required to induce a transition from parallel to perpendicular configuration . however , for @xmath101 incommensurable with the bulk lamellar spacing ( say @xmath4 ) i.e. halfway between integral lamellar spacings , the free energy of the parallel configuration is maximum and hence a lower electric field can induce a perpendicular transition .
3 . for unstrained films
e.g. @xmath149 , the critical electric field @xmath150 required for a parallel to perpendicular transition scales as @xmath151 @xcite . though the numerical calculations are carried out at discrete values of electric field , we can still verify this behavior : @xmath150 for h = 0.1 and @xmath130 lies between @xmath152 and @xmath153 ( fig .
[ fig10 ] ( a ) ) . taking @xmath154 for @xmath130 , @xmath150 for @xmath155 can be calculated to be @xmath156 which lies between the values of @xmath157 and @xmath158 in fig . [
fig10](b ) .
similar trend can be verified for @xmath137 in fig . [ fig10](a ) and ( b ) .
4 . for higher substrate interaction strengths , i.e @xmath159 , the range of stability of parallel configuration increases with decreasing film thickness ( fig .
[ fig10](a ) , ( b ) , ( d ) , ( e ) ) . for @xmath98 ,
the parallel configuration is more stable .
mixed morphologies are stable only for thicker films ( @xmath160 ) and for higher substrate interaction strength , @xmath161 ( fig .
[ fig10](a ) , ( b ) ) .
interestingly , the electric field required for the transition of parallel to mixed morphology , i.e. the first critical field is independent ( or at most weakly dependent due to the discretized nature of the phase diagram ) of substrate interaction strength ( fig . [ fig10](a ) and ( b ) ) .
however , the second critical field i.e. the field required to convert mixed to perpendicular morphology is dependent strongly upon the substrate interaction strength ( approximately linearly ) .
next , we consider the variation of the critical fields with film thickness for unstrained films ( @xmath162 ) for fixed substrate interaction strengths of @xmath163 ( fig . [ fig10 ] ( a ) and ( b ) )
clearly , both the critical fields depend upon the film thickness .
the dependence of the first critical field , though , is stronger than the second critical field .
interesting morphologies arise at @xmath164 where the film thickness is incommensurable with the lamella period . at low substrate affinities a usual parallel to perpendicular lamellae transition is observed but at higher affinities a wide range of hybrid structures results . the exact nature ,
though , is not clear in the present study . phase diagram showing the stable arrangement at different magnitudes of electric field @xmath126 and substrate interaction strength @xmath75 for film thickness ( a ) @xmath124 , ( b ) @xmath165 for permittivity values of @xmath166 and @xmath167 . ] the effect of increasing the dielectric contrast i.e. @xmath168 on the final configuration , will now be discussed .
increasing the contrast between the two blocks implies that the material is more responsive to an applied electric field and , as a consequence , if @xmath169 is large , the resulting phase diagram is governed by the relative mismatch . in the studies
so far , we consider @xmath170 ( @xmath171 and @xmath172 ) , which are close to the values reported by amundson et al .
@xcite ( @xmath173 and @xmath167 ) for ps - pmma copolymer .
if we now increase the permittivity difference @xmath174 by selecting @xmath166 and @xmath167 as reported in ref .
@xcite , for the same copolymer system , an increase in electrostatic free energy contribution is expected for the same magnitude of applied field .
we restrict the discussion to unstrained films ( @xmath162 ) .
the resulting phase diagram is presented in fig .
[ fig11 ] and the corresponding values of the critical electric fields are drastically lowered . according to the analytical calculations
@xcite , this decrease is proportional to @xmath175 . considering the critical electric field to be @xmath154 for @xmath170 and @xmath130 and @xmath99 ( fig .
[ fig10](a ) ) , the critical field on increasing the dielectric contrast to @xmath174 according to above equality ( on holding @xmath101 and @xmath75 constant ) yields a critical value of around @xmath176 which complies well with fig . [ fig11](a ) .
a similar behaviour is retrieved for @xmath155 as well .
the nature of the phase diagram changes dramatically and the region of mixed morphology in the phase diagram is diminished . with enhanced dielectric contrast , the dependence of the critical fields on the substrate interaction strength for a given film thickness becomes rather weak .
this is contrary to the behavior at low dielectric contrast where the second critical field ( mixed to perpendicular ) displayed a strong dependency on substrate interaction strength .
however , the dependency of the critical fields on film thickness for a given substrate interaction strengths is similar to that at low dielectric contrast i.e. both fields depend on the film thickness with the dependency of the second critical field being higher than the first .
we conclude the paper with a critical assessment of the results from the present study to the experiments , analytical and scft calculations .
the most relevant experimental study is due to xu et al .
they studied thin films of varying thickness ( @xmath177 ) @xcite . for film thicknesses less than @xmath178
, substrate interaction was found to be dominant resulting in parallel arrangement , even on application of electric field of @xmath179 .
the substrate induced ordering length in their study was about @xmath180 and the segmental interaction or the segregation was @xmath181 .
the segregation in the present work is relatively weaker than their experiments ( @xmath182 ) and ordering length is around @xmath100 ( with lowest substrate interaction strength ) .
the critical field for lowest substrate interaction from our study is of the order of @xmath183 and @xmath184 for @xmath185 and @xmath186 respectively for a dielectric contrast of @xmath170 .
the second value is actually quiet high and may well exceed the dielectric breakdown of the material and in such cases only parallel ordering would be exhibited .
we remark that in recent experiments , electric field of the order of @xmath187 have been applied @xcite .
however , if we examine the critical field for enhanced dielectric contrast of @xmath174 ( the same as in experiment ) , the values are around @xmath179 and @xmath188 for @xmath185 and @xmath186 respectively , which are well within experimental range . given a higher segregation in experiments , we can expect the critical fields to be higher than that in the present study .
infact using scft matsen @xcite calculated the critical field to be around @xmath189 for the same experimental conditions of xu et al .
( @xmath174 and @xmath181 ) for @xmath190 , though substrate interaction was not explicitly considered .
moreover if we compare the critical field for @xmath182 from their work , the critical field would turn about to be roughly @xmath191 for dielectric contrast of @xmath174 and @xmath185(fig 5(a ) in ref .
@xcite ) .
we additionally compare our findings to that of lin et al .
@xcite who studied sphere to cylinder transition . using scft calculations , a complete phase diagram was calculated .
their segregation also corresponds to @xmath182 .
the present work is thus complementary to their study . for weak substrate interaction , the critical field ( maximum value ) calculated by them
is around @xmath192 and @xmath193 for @xmath185 and @xmath186 respectively ( fig . 7 in ref .
@xcite ) . considering that sphere to cylinder transition generally takes place at field strength lower than parallel to perpendicular lamella transition , our values of @xmath179 and @xmath188 are quite in agreement with scft calculations .
a similar comparison can also be made at higher substrate strengths .
the calculated phase diagrams are similar in spirit to that by lyakhova et al .
@xcite . in both studies
, the phase diagram is obtained from dynamic microstructure evolution rather than static calculations @xcite .
the authors investigated parallel to perpendicular transition of lamellar morphology using dynamic scft coupled to perturbed solution of maxwell equation for thin films of @xmath194 and segregation of @xmath195 .
mixed morphologies were however not observed in that study . in the present study ,
the system was allowed to evolve from a disordered state under combined electric and substrate field , whilst in the study of lyakhova et al .
, electric field was applied to well developed microstructures .
possible difference can arise because of the initial level of ordering .
our results are qualitatively similar to the results of lin et al .
@xcite who observed the presence of mixed phases in film thickness as low as @xmath186 in cylinder forming systems at a similar segregation .
we next compare our phase diagram to the analytical calculation of tsori et al . @xcite .
the authors computed the phase diagrams both in weak and strong segregation regime .
the results presented correspond to an intermediate regime . though our results are closer to wsl , the phase diagram presented in fig .
[ fig10 ] is similar to the analytical calculation of tsori et .
al in ssl ( fig . 8 and 9 in ref .
we remark that the two phase diagrams correspond to two different segregation regimes and are based on different assumptions .
phase diagram calculated by tsori et al
. corresponds to ssl and is based on the assumption of finite surface ordering length and high dielectric contrast ( @xmath174 as compared to our @xmath170 ) . in wsl , they assumed the ordering length to be greater than the film thickness and mixed morphology was not considered . in the present study , mixed morphology is observed inspite of the ordering length being lower than the film thickness .
we believe that the consideration of finite surface ordering length in context of wsl ( and in the occurence of mixed morphology in particular ) might have been an over assumption .
moreover , only a single intermediate phase i.e. a mixed lamellae morphology was considered .
the results of the present study , however , indicates the presence of other intermediate or hybrid structures , thus altering the phase diagram significantly from analytical theories .
we conclude by briefly discussing the influence of wall interaction characteristics on the equilibrium morphologies . in the present work ,
we have restricted the study to symmetric substrate interaction .
however , in principle two additional cases are possible
. the substrate can be ( i ) antisymmetric i.e. both walls attract different monomers with same strength or ( ii ) asymmetric i.e. both walls have preference towards same or different monomers , but possess different interaction strength .
we believe that the consideration of antisymmetric case may not lead to any new geometries other than the ones reported here .
only the region of their respective occurrence in the phase diagram might change , given that film thickness corresponding to integral number of lamellae spacing would then be the frustrated state and half - integral , the natural state . on the contrary , asymmetric interaction can potentially engender an additional type of mixed morphology , where the system adopts a parallel configuration on one side ( where interaction strength is comparatively higher ) and perpendicular configuration on the other ( where interaction strength is weaker ) .
however , it remains to be seen if the competing electric field can stabilize other morphologies , for instance , cylindrical that has been reported in the present study . to summarize
, we have studied the morphology evolution of a symmetric diblock copolymer under competing substrate interaction and electric field using a coupled ohta - kawasaki functional and maxwell equation . by solving the full maxwell equation
, we do not assume weak dielectric inhomogeneity , making the model equally applicable irrespective of segregation . a good agreement with the analytical and scft calculations ,
amply demonstrates the predictive capability of the proposed model .
a distinct advantage of coarse graining is the accessability to large scale simulation , especially in three dimension .
moreover , in 3d the nature of mixed / hybrid morphologies are well defined and such a simple classification ( parallel , perpendicular and mixed lamellae ) may not be suffice .
our 2d results does point out that in the incommensurate films in the regime @xmath196 , this interplay of substrate , confinement and electric field leads to rich hybrid structures and even the occurence of a parallel lamellae to perpendicular cylinder transition .
infact in recent experiment on gyroid forming copolymers by crossland et al .
@xcite a large number of coexisting morphologies were observed at low dielectric contrast .
we present some preliminary 3d results in fig .
[ fig12 ] and [ fig13 ] which corroborates the findings of our present 2d study .
the exact nature of these hybrid structures and their transition in electric field will be communicated shortly .
62ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop _ _ , vol .
( , ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) | in the present work , we study morphologies arising due to competing substrate interaction , electric field and confinement effects on a symmetric diblock copolymer . we employ a coarse grained non - local cahn - hilliard phenomenological model taking into account the appropriate contributions of substrate interaction and electrostatic field . the proposed model couples the ohta - kawasaki functional with maxwell equation of electrostatics , thus alleviating the need for any approximate solution used in previous studies .
we calculate the phase diagram in electric field - substrate strength space for different film thicknesses .
in addition to identifying the presence of parallel , perpendicular and mixed lamellae phases similar to analytical calculations , we also find a region in the phase diagram where hybrid morphologies ( combination of two phases ) coexist .
these hybrid morphologies arise either solely due to substrate affinity and confinement or are induced due to the applied electric field .
the dependence of the critical fields for transition between the various phases on substrate strength , film thickness and dielectric contrast is discussed .
some preliminary 3d results are also presented to corroborate the presence of hybrid morphologies . |
quantum compass model @xcite has recently attracted great interests @xcite .
it was originally proposed as a simplified model to describe some mott insulators with orbit degeneracy described by a pseudospin . in particular
, compass model describes a system where the anisotropy of the spin coupling is related to the orientation of bonds .
more recently , it was also proposed as a realistic model to generate protected qubits @xcite .
it has been argued that the eigenstates of the quantum compass model are two and only two fold degenerate @xcite .
the two - fold degenerate ground state can be an implementation of a protected qubit if it is separated from the low energy excitations by a finite gap .
symmetry will protect weak noise destroying the degeneracy . based on the protected qubits , which has high quality factor , a
scalable and error - free schema of quantum computation can be designed@xcite .
the two - fold degenerate state has been shown to be gapped from low energy excitations based on the results calculated in a small size system@xcite .
however , the results can not be extended to large size systems .
the results from both spin wave study and exact diagonalization have suggested that the system develops a spontaneously symmetry - broken state in the thermodynamic limit@xcite . in the spin - wave study
, the hamiltonian is expanded around the uniform classical ground state up to the first order of @xmath1 .
it is found that there is a directional ordering of the ground state@xcite . from finite size diagonalization of samples with size up to @xmath2
, it is shown that on clusters of dimension @xmath3 , the low - energy spectrum consists of @xmath4 states which collapse onto each other exponentially fast with @xmath5 . at the symmetric point of @xmath0 ,
@xmath6 states collapse exponentially fast with @xmath5 onto the ground state . from both of the spin - wave analysis and exact diagonalization
, a first order phase transition at the symmetric point seems most favorable .
however , the spin - wave analysis ignores very important fluctuations while the exact diagonalization is limited by small sample size . in this work ,
we first show that the spin-@xmath7 quantum compass model on a square lattice can be exactly mapped to a fermionic model with local density interactions .
normally , after performing a jordan - wigner transformation , one expects a non - local gauge interaction between fermions in the fermionic hamiltonian@xcite .
however , due to the special structure of both the spin interactions and the lattice , we show that the gauge interaction for the compass model is absent , which allows us to apply conventional approximation techniques developed for electron systems to analyze the original spin model .
our approximation method automatically takes into account the most important fluctuations , those perpendicular to the ordering direction .
the remaining fluctuations can also be studied in the perturbative approach .
it is shown that our conclusion is robust against the perturbative corrections .
our results support that a first order phase transition happens at @xmath8 between two different states with spin ordering along either @xmath9 or @xmath10 directions .
the compass model appears rather simple on the first look , @xmath11 where @xmath12 is a two dimensional coordinate .
the sign of @xmath13 and @xmath14 is not important .
we can always introduce a transformation to bring them into @xmath15 .
we assume temperature @xmath16 throughout this work and set @xmath17 for simplicity .
this model has some interesting symmetries , which have been discussed in details in literature @xcite .
we first recall two types of symmetry generators of this model@xcite @xmath18 it has been shown that the symmetry of this model leads to a one - dimension type of behavior and directional ordering @xcite .
it is also shown @xcite that this model is dual to recently studied models of @xmath19 superconducting arrays@xcite , which also show the effect of dimensional reduction @xcite . with the help of the symmetry operators defined in eq.([symmetry - pq ] ) , we find that it is possible to fermionize the full compass model .
the resulted fermionic model has bcs - type pairing along one direction and nearest neighboring interactions along the other direction .
the jordan - winger transformation can be defined as @xmath20\right)c^\dag_{i , j}\\ s^z_{i , j}&=&c^\dag_{i , j}c_{i , j}^{}-\frac{1}{2},\end{aligned}\ ] ] where @xmath21 annihilates a fermion at site @xmath12 .
the original compass model eq.([h - full ] ) is transformed into @xmath22 \end{aligned}\ ] ] up to a constant .
because the interaction along @xmath10 direction is quartic , the string - type interaction or gauge interaction disappears .
the symmetries defined in eq.([symmetry - pq ] ) also have profound implications on the possible ordering of this model .
based on the fact that @xmath23=[q_j , q_{j'}]=[p_i , q_jq_{j'}]=[p_ip_{i'},q_j]=0 $ ] and @xmath24\neq 0 $ ] , it is found that each eigenstate of this model is two and only two fold degenerate @xcite .
this leads to the observation that any finite system can not have an ordering that is characterized by some finite mean - field expectation value of either or both @xmath25 and @xmath26 .
for instance , let us assume the ground state @xmath27 of a @xmath3 system has finite @xmath26 .
one can thus apply @xmath28 onto @xmath27 and obtain @xmath4 degenerate ground state , which is of course inconsistent with the above result of double - degeneracy .
however , ordering is still possible when the system goes to the thermodynamic limit where spontaneous symmetry breaking happens .
the excitation gap that separates the true ground state and other @xmath29 low energy excitation states collapses exponentially as the system size goes to infinity@xcite . in this case , the spontaneously broken symmetries are the @xmath30 symmetries of the one - dimensional ising chain along the ordering direction .
let us consider a system @xmath31 and let both @xmath32 and @xmath33 increase to infinity .
if the ordering is along the @xmath10 direction , one should observe @xmath34 states collapse onto the true ground state and the gap vanishes exponentially , namely @xmath35 with some length scale @xmath36 .
since the spontaneously broken symmetries are a large number of copies of discrete symmetry @xmath30 , there is no corresponding goldstein mode .
furthermore , one can also effectively view this spontaneously symmetry breaking as the breaking of a set of local @xmath30 symmetries in a one - dimensional problem , which is along @xmath9 if the ordering is along @xmath10 .
this does not violates the elitzur s theorem @xcite since in this case the effective local @xmath30 symmetry is realized by a infinitely long ising chain and thus the energy barrier to restore the symmetry is still infinite
. we shall now consider the phase diagram in the thermodynamic limit .
we start from two extreme cases which are trivial to solve .
we then study how the system evolves from one extreme to the other .
it turns out that by studying this question it is possible to obtain the phase diagram as we shall show below .
let us first start from the simplest limit @xmath37 , where the model reduces to decoupled ising chains .
the ground state is magnetically ordered such that @xmath38 up to an overall flip of all spins of each ising chain .
this ordered state thus has @xmath4 degeneracy for a @xmath3 system .
similarly , at the limit of @xmath39 , the spins are ordered along @xmath9 direction .
we are thus interested in how the spins rotate from @xmath10 direction to @xmath9 direction as the ratio @xmath40 changes from @xmath41 to @xmath42 and what is the possible ground state at @xmath0 . in very general , there are four possible scenarios .
the first possibility is that there is no phase transition .
the spins rotate continuously from @xmath10 to @xmath9 through a crossover .
the second scenario corresponds to two second order phase transitions with coexisting of ordering along both @xmath10 and @xmath9 directions .
the third possibility is two second order phase transitions with disordered state in between .
the fourth one is that the spins suddenly change from @xmath9 direction to @xmath10 direction , i.e. , a first order phase transition happens at @xmath43 . to find out which of the four possible scenarios happens
, we start with limit @xmath37 , where the system is ordered along @xmath10 direction , and increase @xmath13 ( or decrease the ratio @xmath44 ) .
as @xmath13 is increased , the flipping of two spins on adjacent chains is introduced .
this process tends to reduce the ordered moment along the @xmath10-direction . on the other hand , the ordering along @xmath10 direction acts as an effective transverse field and suppresses the ordering along @xmath9 direction of the spins on adjacent chains .
this effect is described by a 1d ising chain with a transverse field , which has a critical transverse field @xcite .
when @xmath13 is much smaller than @xmath14 , the ordering of spins along @xmath9 direction is fully suppressed by the adjacent spins ordered along @xmath10 direction .
we then ask at what value of @xmath13 the spins start to order along the @xmath9 direction . for the crossover case
, @xmath45 has a finite value as long as @xmath13 is not zero .
for the second scenario , two second order phase transitions with intermediate coexisting state , @xmath46 starts to develop at some value of @xmath47 that is larger than @xmath48 .
for the last two cases , the expectation value of @xmath49 is zero for all @xmath50 , while @xmath51 takes a finite value for all @xmath52 in the first order phase transition scenario . to determine the behavior of @xmath53 and @xmath54 , we self - consistently study the mean - field decoupled hamiltonian @xmath55.\end{aligned}\ ] ] the mean - field decoupling is justified , since we start from the situation where @xmath56 is ordered and want to know when @xmath57 starts to develop a finite expectation value .
this mean field decoupled hamiltonian describes the most important fluctuation in the state with spins ordered along @xmath10 direction .
the ordering along @xmath9 direction is suppressed by the ordering along @xmath10 direction , which behaves as an effective transverse field for @xmath49 . on the other hand , the feedback effect on @xmath58 due to @xmath49 fluctuations is also included in this approach through the self - consistent condition @xmath59 where @xmath60 is the ground state of the mean - field hamiltonian @xmath61 . to minimize the ground state energy ,
it is favorable to have @xmath26 @xmath62-independent , _ i.e. _ , the moment on the same chain is uniform such that the @xmath14 terms are minimized .
the mean - field hamiltonian thus describes a series of decoupled ising chains with an effective transverse magnetic field @xmath63 that is determined self - consistently . to minimize the coupling energy along @xmath10-direction
, one would expect @xmath64 is @xmath65-independent . if @xmath66 , we can rotate the spins on @xmath65-th row by an angle @xmath67 around the @xmath9-axis . without losing generality , we can thus consider the case where @xmath68
. in the fermionic language , the mean - field decoupling introduced in eq.([h ] ) corresponds to the decoupling of the interaction term @xmath69 into an effective chemical potential term @xmath70 with @xmath71 this hamiltonian can be diagonalized using bogoliubov transformation @xmath72 where @xmath73 is fourier transformation of @xmath74 .
the @xmath75 and @xmath76 can be solved to yield @xmath77}\\ v_k & = & i\sqrt{\frac{1}{2}\left[1-\frac{(\cos k)/2-\mu}{e_k}\right]}\end{aligned}\ ] ] with @xmath78 .
it follows that the expectation value of @xmath56 is @xmath79 the self - consistent equation thus reads @xmath80 ) .
there is no non - trivial solution for @xmath81 .
( a ) the effective transverse field @xmath82 vs @xmath47 .
( b ) the ordered moment along @xmath10 direction @xmath54 vs @xmath47.,title="fig:",width=230 ] + ) .
there is no non - trivial solution for @xmath81 .
( a ) the effective transverse field @xmath82 vs @xmath47 .
( b ) the ordered moment along @xmath10 direction @xmath54 vs @xmath47.,title="fig:",width=230 ] from the results shown in fig.[fig - solution ] , we know there are nontrivial solutions to the self - consistent equation ( [ eq - selfconsistent ] ) for @xmath83 .
the jump happens at @xmath84 .
below @xmath85 , which is the critical point of the ising chain in a transverse magnetic field , @xmath49 starts to develop a nonzero expectation value .
we notice two important results that can help us determine the phase diagram .
firstly , the critical point @xmath85 happens at a @xmath47 smaller than the symmetry point @xmath86 , below which ordering along @xmath10 direction is no longer a reasonable assumption . for @xmath87 ,
the ordering along @xmath9 direction is instead more favorable .
this implies that @xmath53 can not develop a finite value for @xmath50 .
the first two scenarios , namely crossover scenario and two second - order phase transitions scenario , are thus excluded .
secondly , the self - consistent equation has non - trivial solutions for @xmath88 , which means spins order along @xmath10 direction for @xmath52 . combining these two points ,
we conclude that there is a first order phase transition at @xmath86 . at @xmath89 ,
the spins are ordered along @xmath10 direction while they are along @xmath9 direction for @xmath87 . at @xmath86 ,
the spins are either along @xmath9 or @xmath10 direction since our argument above shows that the ordering at @xmath86 along @xmath9 ( @xmath10 ) direction induces a strong enough transverse field to suppress the ordering along @xmath10 ( @xmath9 ) direction .
this conclusion is consistent with the spin - wave analysis@xcite . in spin wave analysis ,
fluctuations of both directions are taken into account partially . in our approach
, the most important fluctuations , namely the ordering direction of weaker bond , is solved exactly .
although the mean - field approximation has included the most important fluctuations , there is still concerns about the effect of the ignored fluctuations . to further study the effect of fluctuations , we calculate the energy gap by perturbation .
the perturbation comes from the interaction terms that are ignored in the mean - field hamiltonian eq.([h ] ) @xmath90 in terms of feynman diagrams , there are hence two kinds of vertices : two - particle interaction vertex and external field interaction .
let us first work out the feynman rules for the two - particle interactions .
we observe that the interactions are between particles on nearest neighboring lines .
the vertex of two - particle interaction thus consists of three parts , the coupling strength , the contribution from @xmath62-th line and the contribution from the neighboring @xmath91-th or @xmath92-th line .
there are four possible types of contributions , which are given in fig.[fig - bcoeff ] with @xmath93 -th line to the vertex . ]
the first line corresponds to @xmath94 . in this section
, we will use @xmath95 to denote the expectation value of operator @xmath96 in the un - perturbed mean - field ground state .
this term can be obtained by contraction from either @xmath97 or @xmath98 .
the coherence factor of @xmath99 in the two - particle interaction is @xmath100 while the one of @xmath101 is @xmath102 .
taking into account of the sign difference between these two contractions , we find @xmath103 .
similarly , we get the results for other three possibilities .
the resulted vertex for two - particle interaction is thus @xmath104 with corresponding feynman diagram sketched in fig.[fig - vertex ] . here
, we have not shown the arrows associated with @xmath105 .
these arrows determine the corresponding subscripts of @xmath106 and @xmath107 .
the corresponding momentum conservation law for the vertex is @xmath108 where the signs of @xmath109 and @xmath110 are the first and second subscripts of @xmath111 .
for instance , the momentum conservation for @xmath112 is @xmath113 .
similarly , we can obtain the vertex for the external field term as illustrated in fig.[fig - vertex - external ] we are now at position to calculate the feynman rule for bubbles .
the clockwise bubble corresponds to @xmath114 with coherence factor @xmath115 while the counterclockwise one is @xmath116 with coherence factor @xmath117 .
the feynman rules for bubble and propagator of particle is shown in fig.[fig - bubble ] a noticeable difference between this propagator and the one in the familiar case of electrons under coulomb repulsion is that we only have the plus sign before the infinitesimal imaginary part .
this is because in the current system , we do nt have a fermi sea as ground state and therefore do not have a hole propagator .
this property of system largely reduces the number of diagrams we need to calculate .
finally , every intermediate 4-momentum should be integrated with a factor of @xmath118 . with all the feynman rules in hands , we are now ready to do perturbative calculations .
the first order correction to the self - energy is given in fig.[fig-1st ] .
the first term is the contribution from the interaction with nearest neighbor lines .
the factor @xmath119 comes from the fact that there are two nearest neighbor lines .
the second term is from the external field " term in eq.([eq - interaction ] ) .
the first order perturbation vanishes in the mean - field decoupling of eq.([h ] ) .
the leading order correction is thus second order perturbation given by the connected amputated graphs as shown in fig.[fig - second - order ] .
+ direction versus @xmath40 .
( b ) the effective field @xmath120 after second - order correction is included .
solid lines are the results with second order perturbation included while dash lines are mean - field results.,title="fig:",width=240 ] + direction versus @xmath40 .
( b ) the effective field @xmath120 after second - order correction is included .
solid lines are the results with second order perturbation included while dash lines are mean - field results.,title="fig:",width=240 ] .
there are five pieces of nonzero second order correction to the single particle propagator .
we detail the calculation of them separately .
the first two diagrams denote the self - energy process involving two particle - particle interactions .
the overall factor of two stems from the fact that a particle on the , say , @xmath62th chain can either interact with the @xmath92th or the @xmath91th chain .
the relative minus in the bracket is there because we invert three ( an odd number ) of the propagators in the second diagram .
direct application of the feynman rules we derived yields @xmath121 where @xmath122 and @xmath123 are the contributions from the first and second pieces of fig.[fig - second - order ] respectively .
the intermediate frequencies @xmath124 and @xmath125 have been integrated .
the third and the fourth diagrams denote self - energy processes involving one particle - particle interaction and one external field perturbation . the factor of 2 stems from the same argument and the overall negative sign can be obtained by working out the contractions .
their analytical expressions @xmath126 and @xmath127 read @xmath128 where again , the intermediate frequency has been integrated .
the last piece of the expression denotes process that involves two external field perturbation .
it is easy to obtain the analytical expression as @xmath129 in this way , we have exhausted all possible self - energy processes to the second order of @xmath14 .
the resulted sum of the diagrams is given by @xmath130 , where @xmath131 is the correction to the propagator of a quasi - particle carrying momentum @xmath132 .
the energy of a particle can be defined as the singularity of its propagator and readily obtained by solving the following equation @xmath133 the energy gap is defined as the lowest energy of excitation , which is numerically calculated and shown in fig.[fig - gap ] .
the second order correction amounts to about @xmath134 of the corrected energy gap itself at the symmetric point @xmath0 . in fig.[fig - sz - j2 ] , we also plot the ordered moment @xmath58 as a function of @xmath40 .
again , the moment is reduced by a small second - order correction . at the symmetric point ,
the effective field is still larger than the critical value @xmath135 below which an ordered moment along @xmath9 direction becomes possible .
we thus conclude that the mean - field results are robust against the second order perturbation .
in conclusion , we show that the compass model can be mapped to a fermion model with local density interaction . through a self - consistent solution of the model
, we argue that there is a first order phase transition at the symmetric point @xmath0 .
this conclusion is consistent with spin - wave analysis and recent numeric computations of the spectrum @xcite . in our approach , the most important fluctuation , namely the fluctuations along the weak coupling direction ,
is taken into account exactly .
the fluctuations along the ordering direction is considered up to second order perturbation .
it is shown that the result of our mean - field approximation is robust against such corrections .
we are grateful for the useful discussions with c. xu , prof .
e. fradkin , and prof .
j. leggett .
this material is based upon work supported by the u.s .
department of energy , division of materials sciences under award no .
defg02 - 91er45439 , through the frederick seitz materials research laboratory at the university of illinois at urbana - champaign and by the national science fundation , division of mathematical physics , under award no.phy-0603759 . | in this work , we show that the quantum compass model on an square lattice can be mapped to a fermionic model with local density interaction . we introduce a mean - field approximation where the most important fluctuations , those perpendicular to the ordering direction , are taken into account exactly .
it is found that the quantum phase transition point at @xmath0 marks a first order phase transition .
we also show that the mean field result is robust against the remaining fluctuation corrections up to the second order . |
recent years , in some results from lattice qcd @xcite , it seems the qcd phase could be changed by a strong magnetic field . by the analysis of some thermodynamic observables ,
it has been found the critical temperature of the crossover region should fall when the magnetic field increases @xcite .
it implies the confinement / deconfinement phase transition or the chiral phase transition @xcite would tend to be induced by a strong magnetic field . with the mit bag model , this result could be reproduced qualitatively @xcite , reflecting the great significance of quark confinement .
furthermore , the approach of large-@xmath0 qcd has already been considered in @xcite . from the analysis of the flavor correction @xmath1 to the pressure , this effect has also been obtained due to the quark degrees of freedom .
on the other hand , gauge / gravity duality or ads / cft has become a framework to understand non - perturbative aspects of strong - coupled quantum field theory @xcite .
motivated by these , we would like to investigate the thermodynamics of the quarks and gluons by using the famous top - down holographic model i.e. the witten - sakai - sugimoto model @xcite , in the presence of a magnetic field with the dynamical flavors . by the underlying string theory ,
this model describes a non - supersymmetric and non - conformal yang - mills theory in 3 + 1 dimension coupled to @xmath2 chiral massless fermions ( quarks ) and adjoint massive matter , as a low energy effective theory . in the original witten - sakai - sugimoto model ,
there are @xmath0 d4-branes compactified on a circle representing the dynamics of gluons , @xmath2 species of massless quarks introduced by putting in @xmath2 pairs of d8 and anti d8-branes ( @xmath3-branes ) . by taking the large @xmath0 limit i.e. @xmath4
, these @xmath0 d4-branes produce a 10d background geometry described by type iia supergravity while the @xmath2 @xmath5-branes are as probes .
accordingly , the fundamental quarks do not have dynamical degrees of freedoms thus are quenched . in order to take into account the dynamics of the fundamental matter and compare it with the results from lattice
qcd , we construct the gravitational solutions by considering the backreaction from the flavor branes with turning on a background magnetic field , at leading order of the veneziano limit .
it means the gravitational solution , as an explicit description of the dual field theory , takes the limit of large number of colors and flavors i.e. @xmath6 , but with fixed @xmath1 , at leading order of the small @xmath1 expansion .
we study the model both in confined and deconfined geometry by re - solving the full action ( bulk fields plus flavor branes ) i.e. taking into account the backreaction from the flavor branes and the magnetic field . and
it also turns out that our solutions describe the configuration with an abelian group as chiral symmetry .
furthermore , the description of deconfinement transition and chiral transition in the witten - sakai - sugimoto model was proposed in @xcite . at zero temperature ,
the bubble ( confined ) solution of the @xmath0 d4-branes is dominant , corresponding the confinement phase in the dual field theory , while the black brane ( deconfined ) solution of the @xmath0 d4-branes arises as the deconfinement phase at high temperature .
thus the phase transition between confinement and deconfinement can be identified as the hawking - page transition between two different background geometries .
we can therefore evaluate the critical temperature by the analysis of the pressure in bubble and black brane background .
the result shows that a confinement / deconfinement transition at @xmath7 arises , where @xmath8 is a mass scale of the mass spectrum .
however , while the bubble solution can be connected to the confinement phase of the dual field theory , this is less clear for the black brane solution because of the mismatched value of the polyakov loops @xcite , which thus makes `` the black d4-brane solution corresponding the deconfinement phase '' may not be strictly rigorous .
nevertheless , we can focus on the chiral transition since the embedded flavor branes take connected / parallel configuration in the bubble / black d4-brane solution respectively , which corresponds to the chirally broken / symmetric phase in the dual field theory . on the other hand ( as we can see in @xcite ) , because of no contributions from the flavors , a constant and flavor - independently critical temperature @xmath9 might be unrealistic .
so in these senses , we would like to study the hawking - page transition of this holographic system by taking into account the backreaction of flavor and the magnetic field , then we are able to compare the resultantly holographic phase diagram with lattice qcd @xcite .
there have been some present works to study the thermodynamics of the quarks and gluons in the presence of a magnetic field by using this model @xcite .
however most of them are discussed in the probe flavor limit .
thanks to the techniques proposed in @xcite and especially used in @xcite , it allows us to consider the backreaction from the flavors in this model corresponding to study the hawking - page transition with flavor and color branes . by employing these techniques , we would like to make some improvements to @xcite with the backreaction from the flavor branes and it could also be treated as a parallel computation to @xcite .
let us outline some technical details in our manuscript .
first , we use the smearing technique @xcite for the flavor branes to construct gravitational solutions as @xcite but in the presence of a magnetic field ( and with zero chemical potential ) .
then in order to preserve the isometries of the original background , we also homogeneously smear a large number @xmath2 of d8-branes on the @xmath10 circle where the @xmath0 d4-branes are wrapped .
as it will be seen , while this configuration simplifies the calculations greatly , we have to solve a set of coupled second order equations of motion of this system . because of the presence of the magnetic field , these equations of motion are all highly non - linear , which are still extremely complicated to solve
. hence we focus on solving these equations in the limit of small magnetic field and small flavor backreaction since it admits analytically magnetic - dependent solutions . and to determine the integration constants in our solution , we furthermore require that the backgrounds must be completely regular in the ir region of the dual field theory . with the presence of the magnetic field , the integration constants could be able to depend on the constant magnetic field .
however , we find it is not enough to determine all the integration constants just by this requirement .
besides , in the uv region , there also is an non - removable divergence unaffected by the presence of the magnetic field , which is due to the landau pole in field theory , reflecting in the running coupling holographically .
last but not least , since the onshell action evaluated by our gravitational solutions is divergent , we need to holographically renormalize the theory in order to study its thermodynamics .
the holographic renormalization of this famous witten - sakai - sugimoto model has been studied with dynamical flavors in a covariant way @xcite ( also see @xcite in the presence of a magnetic field with the approach of probe flavor brane ) .
accordingly , we combine the viewpoints in @xcite to study the holographic renormalization of this model with dynamical flavors and the magnetic field . as the result
, we find if the parameters in the covariant counterterms ( proposed in @xcite ) depend on the magnetic field , they are enough to cancel all the divergences in our calculations . in our probe approximation
the resultantly renormalized pressure is in agreement with @xcite and we can obtain the phase diagram by comparing the confined / deconfined pressure .
our holographic phase diagram shows , the critical temperature decreases when the magnetic field increases , which qualitatively agrees with the lattice qcd results @xcite .
consequently this approach is interesting and could be treated as an improvement to @xcite .
this paper is organized as follows . in the next section 2
, we will give a brief review of the witten - sakai - sugimoto model . and in section 3 , we introduce a magnetic field on the flavor brane and construct the gravitational solution by re - solving the type iia plus flavor brane action , both in confined ( bubble ) case and deconfined ( black brane ) case .
the magnetic - dependent solution is also given in this section . in section 4
, we discuss some physical quantities by imposing the constructed solution with some special constraints in the confined case . in section 5
, it shows the holographic renormalization in our calculation , and we evaluate the renormalized onshell action and the counterterms by our magnetic - dependent solutions . in section 6 , we discuss the holographic phase diagram with the magnetic field in the case of the probe approximation and backreaction respectively , then compare our results with lattice qcd .
discussion and summary are given in the final section .
a non - supersymmetric and non - conformal ( 3 + 1 dimensional ) yang - mills theory was proposed by witten @xcite as the low energy limit of a kaluza - klein ( kk ) reduction of a 4 + 1 dimensional @xmath11 super conformal theory which couples to massless adjoint scalar and fermions .
this theory is the low energy effective theory describing the open string ending on the worldvolume of @xmath0 coincident d4-branes placed in the 10d minkowskian spacetime . by the dimensional reduction ,
the theory is compactified on a circle ( denoted as @xmath10 ) of length @xmath12 . with the choice of boundary conditions for bosons ( periodic b.c ) and fermions ( anti - periodic b.c ) , the massless modes at low energy scale i.e. @xmath13 are the gauge fields of 3 + 1 dimensional @xmath14 yang - mills theory .
the supersymmetry breaks down since the other modes ( including fermions ) get masses @xmath15 . if @xmath16 , where @xmath17 is the string tension and 4d t hooft coupling respectively , the low energy theory could be decoupled from the kaluza - klein modes .
however , as it is known there is not any simple description in the most interesting region @xmath18 in witten s model . as a conjecture by holography
, there should be a dual description in terms of a classical gravity theory on a background arising as the near - horizon limit of sourced @xmath0 d4-branes , and we can therefore obtain many detailed informations in the region of @xmath19 .
such a background produced by @xmath0 d4-branes would have the topology of a product @xmath20 . here
@xmath21 represents the 3 + 1 dimensional spacetime where we live in .
@xmath22 represents the radial direction denoted by the coordinate @xmath23 as the holographic direction , which could be roughly treated as the energy scale of the renormalization group in the dual field theory . in the @xmath24 plane of the subspace
, the confined background looks like a cigar and the size of the @xmath10 circle smoothly shrinks to zero at a finite value @xmath25 of the radial coordinate @xmath23 .
@xmath26 represents the additional dimensions , whose isometry group is @xmath27 identified as a global symmetry group under rotation of the massive kaluza - klein fields .
the theory describes confinement in the dual field theory and the chiral symmetry breaks at zero temperature once it couples to the chiral massless quarks .
it is achieved to add a stack of @xmath2 pairs of suitably @xmath5-branes embedded in the @xmath0 d4-branes background geometry to introduce @xmath2 chiral fundamental massless quarks as @xcite in witten s model .
quarks are in the fundamental representation of color and flavor group since they come from the massless spectrum of the open strings which are stretching between the color and flavor branes . as the flavor @xmath5-branes are probes in this system , their backreaction to the geometric background is neglected .
correspondingly , the fundamental quarks in the dual field theory are in the quenched approximation . besides , the flavor branes offer a @xmath28 symmetry which could be identified as the global flavor symmetry holographically in the dual field theory .
then it is recognized that the flavor branes connect to each other as a u - shape at zero temperature representing chirally broken symmetry automatically . in the confined background
, the geometry is described by the bubble solution of @xmath0 d4-brane with the following metric , @xmath29+\left(\frac{r}{u}\right)^{3/2}\left[\frac{du^{2}}{f\left(u\right)}+u^{2}d\omega_{4}^{2}\right]\nonumber \\
e^{\phi } & = \ g_{s}\left(\frac{u}{r}\right)^{3/4},\ \
\ f_{4}=\frac{3n_{c}}{4\pi}\omega_{4},\ \ \ f\left(u\right)=1-\frac{u_{kk}^{3}}{u^{3}}.\label{eq:(1)}\end{aligned}\ ] ] where @xmath30 is the curvature radius of the background geometry and @xmath31 is the volume form of @xmath26 , @xmath32 and @xmath33 is the string coupling and length respectively .
@xmath34 is dilaton and @xmath35 is the ramond - ramond four form . for the index ,
we have defined @xmath36 . at the scale @xmath8
, the t hooft coupling is defined as @xmath37 in the 4-dimensional theory .
since @xmath38 , the @xmath10 circle shrinks at @xmath39 . in order to omit the conical singularities at @xmath39 , it provides the following relation , @xmath40 here @xmath12 , as the length of the @xmath10 circle , is related to the mass scale @xmath8 by @xmath41 .
there is an alternatively allowed solution which is the black brane solution taking the following metric , @xmath42+\left(\frac{r}{u}\right)^{3/2}\left[\frac{du^{2}}{f_{t}\left(u\right)}+u^{2}d\omega_{4}^{2}\right]\nonumber \\
f_{t}\left(u\right ) & = \ 1-\frac{u_{t}^{3}}{u^{3}}.\label{eq:(3)}\end{aligned}\ ] ] here @xmath43 .
similarly , it provides the following relation with the @xmath44 circle smoothly shrinking to zero at the horizon @xmath45 , @xmath46 therefore , we have the hawking temperature as @xmath47 where @xmath48 is the lengh of @xmath49 in the deconfined geometry .
-branes are located at antipodal points of @xmath10 - circle and always connected , realizing as chirally broken phase . *
right * : in deconfined geometry the flavor @xmath5-branes could be parallel , realizing as chirally symmetric phase . ] in the classical limit , the gravity partition function @xmath50 , which is related to the euclidean onshell action , could be identified as the free energy of the system thermodynamically .
the phase diagram can be obtained by comparing the free energy of the two phases above .
it has turned out the bubble solution is dominant at zero temperature , while the black brane solution of the @xmath0 d4-branes arises at high temperature , which provides the critical temperature of the phase transition as , @xmath51 in this manuscript , we are going to work in the following configuration when the flavor branes are considered . that is , in the confined case , the @xmath5-branes are placed at antipodal points of @xmath10 - circle .
when the temperature increases , the connected position on @xmath5-branes falls into the horizon in the @xmath52 phase as in figure 1 .
it thus turns to the deconfined case and the @xmath5-branes become parallel ( disconnected ) .
accordingly , the chiral symmetry is restored since the flavor symmetry group remains @xmath28 in the configuration of a stake of parallel @xmath5-branes .
so the confined / deconfined or chiral phase transition could be identified as the hawking - page transition of the background with the connected / parallel configuration of the flavor branes .
in the following sections , there would be three relevant and useful coordinates which are @xmath53 . for the reader convenience , the relation between these coordinates and the standard @xmath23 coordinate used in the witten - sakai - sugimoto model ( [ eq:(1 ) ] ) ( [ eq:(3 ) ] )
is summarized as follows , @xmath54 as it will be seen that @xmath55 represents @xmath25 in confined geometry or @xmath56 in deconfined geometry .
we are going to use @xmath57 to replace @xmath58 in ( [ eq : coor transfor ] ) in the deconfined geometry and the explicit definition of @xmath58 or @xmath57 could be found in the following relevant formulas ( in eq.([eq:18 ] ) and eq.([eq:37 ] ) ) . in the witten - sakai - sugimoto model @xcite ,
the flavor @xmath5-branes are treated as the probes embedded in the confined geometry . however , in this subsection we would like to take into account their backreaction to the first order of @xmath1 in the confined case .
so we are going to use the same trick as @xcite , that is to consider a setup where @xmath2 d8-branes are smeared homogeneously along the transverse @xmath10 circle @xcite .
and we will consider the model below the critical temperature ( @xmath59 ) with a background magnetic field on the flavor branes . for @xmath60 ,
the ansatz of the metric in string frame is given as @xcite , @xmath61 where @xmath62 and @xmath63 are functions depended on the holographic coordinate @xmath64 only .
@xmath10 is the compactified coordinate on a circle with the length @xmath41 .
the function @xmath65 are defined as @xmath66 in order to take into account the backreaction of the flavor and the magnetic field , we have to consider the total action in type iia supergravity with the presence of a magnetic field on the flavor branes . the relevant action ( bulk fields plus smeared flavor brane ) is , @xmath67\nonumber \\ & \ \
-\frac{n_{f}t_{8}m_{kk}}{\pi}\int d^{10}x\frac{\sqrt{-\det\left(g+2\pi\alpha^{\prime}f\right)}}{\sqrt{g_{44}}}e^{-\phi}.\label{eq : action with dynamic flavors}\end{aligned}\ ] ] the first part of ( [ eq : action with dynamic flavors ] ) is the action of the bulk fields while the last part arises as the contribution from the dirac - born - infeld ( dbi ) action of @xmath2 d8-branes which are smeared on the transverse @xmath10 circle . here
@xmath68 is related to the 10d newton coupling . in confined geometry
, we consider the antipodal configuration for the flavor branes and put the smeared dbi action on - shell i.e. the embedding coordinate @xmath69 satisfies its equation of motion @xmath70 .
the integration over the radial coordinate has been calculated as two times to account for the presence of two branches at two antipodal points on the @xmath10 .
furthermore , we have turned on a @xmath71 gauge field on the flavor branes which is the dual of an external background magnetic field .
thus as @xcite we set a constant magnetic field @xmath72 , where @xmath73 is dimensionless constant - brane action vanishes since only one conponent of the gauge field strengh is turned on . and
as a consistent solution for the dbi action , it is allowed to set the magnetic field as a constant .
see also @xcite . ] . with the implementation of the ansatz ( [ eq : metric for confined phase ] ) , it yields the following 1d action @xcite @xmath74,\nonumber \\ v & = & 12e^{-2\nu-2\varphi}-q_{c}^{2}e^{4\lambda+\widetilde{\lambda}-4\nu-\varphi}-q_{f}e^{2\lambda-\frac{\widetilde{\lambda}}{2}+2\nu-\frac{3}{2}\varphi}\sqrt{1+b^{2}e^{-4\lambda}},\label{eq:0 + 1 d action in confined phase}\end{aligned}\ ] ] where we have defined @xmath75 note that we are going to use parameter @xmath76 ( or @xmath77 , in the deconfined case ) to weigh the contribution from flavors to the action , and the dot represents the derivatives are w.r.t .
moreover action ( [ eq:0 + 1 d action in confined phase ] ) has to be supported by the zero - energy constraint , @xmath78 which makes the equations of motion from 10d action ( [ eq : action with dynamic flavors ] ) and the effective 1d action ( [ eq:0 + 1 d action in confined phase ] ) coincident if the homogeneous ansatz ( [ eq : metric for confined phase ] ) is adopted .
then the equations of motion from the previous action ( [ eq:0 + 1 d action in confined phase ] ) are as follows ( derivatives are w.r.t .
@xmath64 ) , @xmath79 we have used the definition of ( [ eq : phi defined in confined phase ] ) to replace @xmath65 by the dilaton field @xmath34 . however , we will not attempt to solve equations ( [ eq : eoms in confined phase ] ) exactly , instead , we will focus on the small magnetic field case i.e. keeping only the leading @xmath80 term . on the other hand ,
since our concern is to find a perturbative solution of ( [ eq : eoms to r leading qb ] ) at the first order of the parameter @xmath76 , we write all the relevant functions in ( [ eq : eoms in confined phase ] ) as , @xmath81 then we use the following unflavored solutions as the zeroth order solution ) and ( [ eq : f in conf ] ) are nothing but the compacted d4-brane solution used in the witten - sakai - sugimoto model , expressed in @xmath82 coordinate with the coordinate transformation ( [ eq : coor transfor ] ) . ] , @xmath83 with @xmath84 in order to keep the leading @xmath80 terms , we have the following equations from ( [ eq : eoms in confined phase ] ) for the leading order function @xmath85 in the expansion of ( [ eq : function expand ] ) , ( derivatives are w.r.t .
@xmath82 ) , @xmath86,\nonumber \\
\widetilde{\lambda}_{1}^{\prime\prime}-\frac{9}{2}\frac{e^{-3r}}{\left(1-e^{-3r}\right)^{2}}\left(4\lambda_{1}+\widetilde{\lambda}_{1}-\phi_{1}\right ) & = & -\frac{1}{4}\frac{e^{-3r/2}}{\left(1-e^{-3r}\right)^{13/6}}\left[1+\frac{1}{2}q_{b}^{2}\left(1-e^{-3r}\right)\right],\nonumber \\
\phi_{1}^{\prime\prime}-\frac{9}{2}\frac{e^{-3r}}{\left(1-e^{-3r}\right)^{2}}\left(4\lambda_{1}+\widetilde{\lambda}_{1}-\phi_{1}\right ) & = & \frac{1}{4}\frac{e^{-3r/2}}{\left(1-e^{-3r}\right)^{13/6}}\left[5+\frac{1}{2}q_{b}^{2}\left(1-e^{-3r}\right)\right],\nonumber \\
\nu_{1}^{\prime\prime}-\frac{3}{2}\frac{e^{-3r}}{\left(1-e^{-3r}\right)^{2}}\left(12\nu_{1}+4\lambda_{1}-5\phi_{1}+\widetilde{\lambda}_{1}\right ) & = & \frac{1}{4}\frac{e^{-3r/2}}{\left(1-e^{-3r}\right)^{13/6}}\left[1+\frac{1}{2}q_{b}^{2}\left(1-e^{-3r}\right)\right].\nonumber \\ \label{eq : eoms to r leading qb}\end{aligned}\ ] ] here @xmath87 and we have assumed @xmath88 .
@xmath89 is the t hooft coupling constant which should be fixed .
other relevant parameters are defined as @xmath90 with the equations in ( [ eq : eoms to r leading qb ] ) , we find that , @xmath91 where @xmath92 and @xmath93 are integration constants and @xmath94 are two particular functions which satisfy @xmath95 the equations in ( [ eq : eoms to r leading qb ] ) would be quite easy to solve after a re - combination and the definition of @xmath96 , it yields an equation for @xmath97 which is which is used in ( [ eq : solution for confined phase - 1 ] ) . ] , @xmath98 so we have the following solution expressed in terms of generalized hypergeometric functions as , @xmath99 and the relevant functions in ( [ eq : solution for confined phase - 1 ] ) are @xmath100,\nonumber \\ y & = & c_{2}-\coth\left(\frac{3}{2}r\right)\left(c_{1}+c_{2}\left(1+\frac{3}{2}r\right)\right)+z+q_{b}^{2}w,\nonumber \\ z & = & -\frac{e^{-9r/2}\left(1+e^{-3r}\right)\left(9e^{3r}\ _ { 3}f_{2}\left(\frac{1}{2},\frac{1}{2},\frac{19}{6};\frac{3}{2},\frac{3}{2};e^{-3r}\right)+\ _ { 3}f_{2}\left(\frac{3}{2},\frac{3}{2},\frac{19}{6};\frac{5}{2},\frac{5}{2};e^{-3r}\right)\right)}{162\left(1-e^{-3r}\right)}\nonumber \\ & & -\frac{8e^{-3r/2}\left(3 + 10e^{-3r}\right)\ _ { 2}f_{1}\left(\frac{1}{6},\frac{1}{2},\frac{3}{2},e^{-3r}\right)}{819\left(1-e^{-3r}\right)}+\frac{e^{-15r/2}\left(38e^{3r}+8e^{6r}-40\right)}{273\left(1-e^{-3r}\right)^{13/6}},\nonumber \\ w & = & \frac{e^{-3r/2}\left(2e^{-3r}+1\right)}{7\left(1-e^{-3r}\right)^{7/6}}-\frac{e^{-3r/2}\left(4e^{-3r}+3\right)\
_ { 2}f_{1}\left(\frac{1}{6},\frac{1}{2},\frac{1}{2},e^{-3r}\right)}{21\left(1-e^{-3r}\right)}\nonumber \\ & & -\frac{e^{-3r/2}\left(1+e^{-3r}\right)\left(e^{-6r}\ _ { 3}f_{2}\left(\frac{5}{2},\frac{5}{2},\frac{19}{6};\frac{7}{2},\frac{7}{2};e^{-3r}\right)-25\ _ { 3}f_{2}\left(\frac{1}{2},\frac{1}{2},\frac{19}{6};\frac{3}{2},\frac{3}{2};e^{-3r}\right)\right)}{300\left(1-e^{-3r}\right)},\nonumber \\ v & = & y-6m_{2}+3\coth\left(\frac{3}{2}r\right)\left(m_{2}\left(3r+2\right)+m_{1}\right)-5z+\frac{q_{b}^{2}}{3}w.\label{eq : solution for confined phase - 2}\end{aligned}\ ] ] here @xmath101 are eight integration constants and some of them could be determined by some physical requirements .
for example , the zero - energy constraint ( [ eq : zero - energy constraint ] ) provides a condition at @xmath25 , to the first order in @xmath76 , which is @xmath102 other constraints for the integration constants in ( [ eq : solution for confined phase - 1 ] ) ( [ eq : solution for confined phase - 2 ] ) would arise by analysing the asymptotics of this solution . since our solution is a perturbation to the zero - th order solution ( [ eq:0th solution for confined phase ] )
, it should be regularity at the tip of the ( @xmath103 ) cigar which corresponds to the limit of @xmath104 ( i.e. it gives ir behavior ) . as a comparison with @xcite
, we work in @xmath105 coordinate and obtain the following ir asymptotics ( @xmath104 i.e. @xmath106 ) , @xmath107 accordingly , it yields the following constraints for the integration constants , @xmath108 note that ( [ eq : constant for confined ] ) satisfies ( [ eq : zero - energy constraint - 2 ] ) automatically . and the uv behavior of functions are given as follows ( @xmath109 i.e. @xmath110 ) , @xmath111 with @xmath112,\nonumber \\ \label{eq : x->1 confined - 2}\end{aligned}\ ] ] the sub - leading terms in ( [ eq : x->1 confined - 1 ] ) , diverging as @xmath113 , do not depend on any integration constants , which are same as in @xcite and could be interpreted as the dual of the `` universal '' terms . in the uv asymptotics ,
the combinations of the appearing integration constants may be interpreted as corresponding to some gauge invariant operators , however it is less clear about what the combinations of these functions correspond to gauge invariant operators
. nevertheless , in order to omit the sources or vevs of the dual operators , at least to switch off the most divergent terms in ( [ eq : x->1 confined - 2 ] ) , we impose the prudent condition as @xcite , @xmath114 and we do not have any more constraints on the integration constants appearing in our solution , thus the integration constants @xmath115 could not be determined and we have to keep them generic . as it can be seen from ( [ eq
: constants - 2 confined ] ) , these integration constants relate to @xmath116 which is another constant as the input of our theory .
although this is a bit different from the case without the magnetic field in @xcite , a possibly special choice of @xmath115 may remain as @xcite , which is @xmath117 since our solution is based on the expansion of small @xmath116 , we can , for example , fix @xmath118 in ( [ eq : spectial choice for const ] ) and look for the relations between @xmath119 and @xmath116 if necessary . in the case without the magnetic field @xcite , a function @xmath120 is defined as @xmath121 and its equation of motion is @xmath122 .
however in our manuscript , by keeping the leading @xmath80 terms , we can obtain the equation of motion for @xmath120 from ( [ eq : eoms in confined phase ] ) as @xmath123 , which makes the arguments for the special choice of @xmath115 in @xcite less valuable .
however , as we consider the small magnetic field , the deformation of @xmath120may also be small i.e. @xmath124 , in this sense , the special choice ( [ eq : spectial choice for const ] ) might also be allowed in the case of small magnetic field
. however these are not strictly necessary . ] .
the deconfined background geometry of this model in unflavored case corresponds to the black d4-brane solution .
the circle @xmath10 never shrinks while the euclideanized temporal circle shrinks at @xmath45 .
the flavor branes take the position at @xmath125 and the configuration of a stack of parallel @xmath5-branes is recognized as the chirally symmetric phase in the dual field theory . similarly as the confined case , we turn on a constant @xmath71 gauge field strength as a background magnetic field on the flavor branes and consider two stacks of flavor branes smeared on the @xmath10 circle .
the relevant action ( with the flavor branes putting onshell ) reads as ( [ eq : action with dynamic flavors ] ) .
we use the following ansatz for the metric in string frame as , @xmath126 where @xmath127 and we also adopt the ansatz for the gauge field strength as @xmath72 as the confined case . here
@xmath73 also represents a dimensionless constant . inserting the ansatz ( [ eq : metric in deconfined phase ] ) and the magnetic field into ( [ eq : action with dynamic flavors ] )
, it yields the following 1d action , @xmath128,\nonumber \\ v & = & 12e^{-2\nu-2\varphi}-q_{c}^{2}e^{3\lambda+\lambda_{s}+\widetilde{\lambda}-4\nu-\varphi}-q_{f}e^{\frac{3}{2}\lambda-\frac{1}{2}\lambda_{s}+\frac{1}{2}\widetilde{\lambda}+2\nu-\frac{3}{2}\varphi}\sqrt{1+b^{2}e^{-4\lambda}}.\label{eq:1d action for deconfined phase}\end{aligned}\ ] ] similarly , this action ( [ eq:1d action for deconfined phase ] ) should also be supported by the zero - energy constraint ( [ eq : zero - energy constraint ] ) .
then we can obtain the equations of motion as ( derivatives are w.r.t .
@xmath64 ) @xmath129 since we are going to search for a perturbative solution in the first order of @xmath1 , we choose the zero - th order solution as the unflavored solution for deconfined case , which is @xmath130 where we have defined @xmath131,\end{aligned}\ ] ] and @xmath132 then we expand all the fields as what we have done in the confined case , @xmath133 with @xmath134 here the relation between @xmath135 and @xmath136 from zero - th order solution has been imposed .
and we have required that @xmath137 at the phase transition which thus suggests a definition of running coupling as @xcite .
then the equations of motion for the leading order functions used in the metric are ( derivatives are w.r.t .
@xmath82 ) , @xmath138 we will also focus on the case of small magnetic field instead of solving ( [ eq : eoms for deconfined phase ] ) exactly in @xmath116 , i.e. keeping @xmath139 terms by an expansion .
so in a word , we need to solve the following equations , @xmath140 with the similar tricks used for the confined case , we thus have the solution as , @xmath141 where the functions in ( [ eq : solution for deconfined phase - 1 ] ) are given as @xmath142+\sqrt{3}\tan^{-1}\left[\frac{2\left(1-e^{-3r}\right)^{1/6}+1}{\sqrt{3}}\right]\nonumber \\ & & 2\tanh^{-1}\left[\left(1-e^{-3r}\right)^{1/6}\right]-\coth^{-1}\left[\left(1-e^{-3r}\right)^{1/6}+\frac{1}{\left(1-e^{-3r}\right)^{1/6}}\right],\nonumber \\ h\left(r\right ) & = & -\frac{4}{\left(1-e^{-3r}\right)^{1/6}}+\frac{2}{3}f\left(r\right),\nonumber \\
y\left(r\right ) & = & c_{2}-\coth\left(\frac{3r}{2}\right)\left[c_{2}\left(\frac{3}{2}r+1\right)+c_{1}\right]+q_{b}^{2}i\left(r\right)+z\left(r\right),\nonumber \\ z\left(r\right ) & = & \frac{3e^{3r}\left(1-e^{-3r}\right)^{5/6}-\frac{\sqrt{3}}{2}\left(e^{3r}+1\right)\left[\tan^{-1}\left(\frac{2\left(1-e^{-3r}\right)^{1/6}-1}{\sqrt{3}}\right)+\tan^{-1}\left(\frac{2\left(1-e^{-3r}\right)^{1/6}+1}{\sqrt{3}}\right)\right]}{546\left(e^{3r}-1\right)}\nonumber \\ & & + \frac{1}{2}\left(e^{3r}+1\right)\frac{2\tanh^{-1}\left[\left(1-e^{-3r}\right)^{1/6}\right]+\coth^{-1}\left[\left(1-e^{-3r}\right)^{1/6}+\frac{1}{\left(1-e^{-3r}\right)^{1/6}}\right]}{546\left(e^{3r}-1\right)},\nonumber \\
i\left(r\right ) & = & -\frac{\left(7e^{-3r}+3\right)}{7\left(1-e^{-3r}\right)^{7/6}}+\frac{5\left(1+e^{-3r}\right)}{42\left(1-e^{-3r}\right)}f\left(r\right),\nonumber \\
w\left(r\right ) & = & 2m_{2}-\left[m_{1}+\left(3r+2\right)m_{2}\right]\coth\left(\frac{3r}{2}\right)+\frac{1}{12}\left(a_{1}-5a_{2}+a_{3}+b_{1}r-5b_{2}r+b_{3}r\right)\nonumber \\ & & + \frac{5}{3}z\left(r\right)-q_{b}^{2}i\left(r\right)+y\left(r\right).\label{eq : solutions for deconfined phase - 2}\end{aligned}\ ] ] the integration constans are represented by @xmath143 . and the zero - energy condition ( [ eq : zero - energy constraint ] ) in the case of small @xmath116 thus is @xmath144 where @xmath145 notice that ( [ eq : zero energy condition ] ) would be satisfied with the leading order solution if @xmath146 the near horizon ( i.e. @xmath106 ) behavior of the relevant functions are given as follows , @xmath147 furthermore , we require that the solution is regular at the tip of the euclidean cigar , it thus leads the following constraints @xmath148 notice that ( [ eq : constraint for def.constant - 1 ] ) fulfills the zero - energy constraint ( [ eq : zero energy condition - 2 ] ) automatically as well . and the uv behavior ( i.e. @xmath149 ) of these functions is , @xmath150+\mathcal{o}\left(\left(1-x\right)^{1/6}\right),\nonumber \\ \lambda_{s1 } & = & -\frac{c_{1}+c_{2}}{1-x}+\frac{29}{455\left(2\right)^{1/6}\left(1-x\right)^{1/6}}+\frac{1}{4}\left[-3a_{1}-a_{2}+a_{3}+2\left(c_{1}+c_{2}\right)\right]+\mathcal{o}\left(\left(1-x\right)^{1/6}\right),\nonumber \\
\widetilde{\lambda}_{1 } & = & -\frac{c_{1}+c_{2}}{1-x}+\frac{101}{455\left(2\right)^{1/6}\left(1-x\right)^{1/6}}+\frac{1}{4}\left[a_{1}-a_{2}-3a_{3}+2\left(c_{1}+c_{2}\right)\right]+\mathcal{o}\left(\left(1-x\right)^{1/6}\right),\nonumber \\
\phi_{1 } & = & -\frac{c_{1}+c_{2}}{1-x}+\frac{361}{455\left(2\right)^{1/6}\left(1-x\right)^{1/6}}+\frac{1}{4}\left[a_{1}-5a_{2}+a_{3}+2\left(c_{1}+c_{2}\right)\right]+\mathcal{o}\left(\left(1-x\right)^{1/6}\right),\nonumber \\
\nu_{1 } & = & -\frac{m_{1}+2m_{2}}{1-x}+\frac{25}{91\left(2\right)^{1/6}\left(1-x\right)^{1/6}}+\frac{1}{12}\left[a_{1}-5a_{2}+a_{3}+6m_{1}\right]+\mathcal{o}\left(\left(1-x\right)^{1/6}\right).\nonumber \\\end{aligned}\ ] ] to eliminate the leading divergences as discussed in the confined case , we impose @xmath151 then we do not have any more constraints for other integration constants , thus we have to keep @xmath152 and @xmath153 generic .
nevertheless a possible choice for @xmath152 and @xmath154 with small magnetic field might be ( same as @xcite ) , @xmath155 however , we have to keep in mind that ( [ eq:47 ] ) is also not strictly necessary .
in this section , we will study some holographically physical effects in hadronic physics by using our magnetic - dependent backreaction solution in confined case ( [ eq : solution for confined phase - 1 ] ) ( [ eq : solution for confined phase - 2 ] ) . to begin with ,
since the ( @xmath156 ) cigar has to close smoothly at the tip ( @xmath104 ) , the relation between the parameter @xmath55 and @xmath8 is modified by the backreaction from the flavor and magnetic field .
therefore we have @xmath157.\label{eq : modified mkk}\ ] ] if using the special choice ( [ eq : spectial choice for const ] ) as @xcite , we obtain @xmath158 . obviously , with this choice , the length of the @xmath10 circle becomes larger as the magnetic field increases . for the reader convenience ,
we also give the relation between the parameter @xmath56 , @xmath159 and hawking temperature @xmath160 in the deconfined case , @xmath161,\label{eq : modified t deconfined}\ ] ] as the metric has to be regular at the horizon of the euclideanized black hole as well . notice that we have to keep in mind all the discussions in this section would not be strictly rigorous once the special choice ( [ eq : spectial choice for const ] ) for the undetermined integration constants is imposed . since all our resutls should definitely return to @xcite
if turning off the magnetic field , we assume ( [ eq : spectial choice for const ] ) ( from @xcite , i.e. the non - magnetic case ) is a simple choice for the undetermined integration constants .
absolutely this is not necessary or strict in our magnetic case .
but because of the lack of the geometric constraints for our gravitational solution and the less clear relation between the integration constants and the magnetic field , some integration constants generic are not determined in fact .
so we can not conclude or compare anything with @xcite if keeping all the undetermined constants generic . according to these
, we therefore impose the special choice ( [ eq : spectial choice for const ] ) throughout the calculations in the following subsections .
consequently our results in this section might not be strictly conclusive but they are good comparisons with @xcite as a check . ] . by examining a d4-brane as the probe wrapped on the @xmath10 circle , we obtain the running gauge coupling @xcite (
the formulas are expressed in the coordinate of @xmath162 . )
@xmath163.\label{eq:54}\ ] ] according to the uv behavior ( @xmath149 ) ( [ eq : x->1 confined - 1 ] ) of the functions , we thus obtain the formula of the running coupling which remains as @xcite , @xmath164.\label{eq:55}\ ] ] obviously , this formula is independent on the presence of the magnetic field which seems different from qft / qcd approach as @xcite but in agreement with @xcite .
technically , ( [ eq:55 ] ) corresponds to the condition ( [ eq : constants - 2 confined ] ) we have chosen . in ( [ eq : x->1 confined - 2 ] ) we have omitted the most divergent terms by imposing ( [ eq : constants - 2 confined ] ) to turn off the sources or vevs of some gauge invariant operators in the dual field theory although some details about the holographic correspondence here are also less clear .
so the surviving divergences in ( [ eq : x->1 confined - 2 ] ) are all independent on the integration constants , which thus yields a integration - constant - independent divergence in ( [ eq:55 ] ) by ( [ eq:54 ] ) . in this sense
our ( [ eq:55 ] ) is same as @xcite since we have chosen the same boundary conditions for the gravitational solution while the gravitational solution itself is actually different . on the other hand ( [ eq:55 ] ) signals a landau pole since the coupling constant tends to diverge in the uv limit ( i.e. @xmath149 ) which strongly differs from qcd in fact .
there might be a simple interpretation about the appearance of the landau pole . as it is known the background of this model is witten s geometry @xcite at the limit @xmath165 . in our backreaction case
, we could require @xmath0 is large but not infinity and @xmath166 fixed .
accordingly , the background geometry is actually 11d ( @xmath167 ) while the 11th direction is compacted on a cycle with a very small size ( as some energy scales in the dual field theory ) .
therefore the dual field theory could be conformal upon this energy scale @xcite .
so it is possible to generate a landau pole by adding flavors to a cft . besides ( [ eq:55 ] ) only shows the the uv behavior ( @xmath149 ) of the running coupling , but basically we can obtain the complete relation between the running coupling and the magnetic field by using ( [ eq:54 ] ) .
the behavior of @xmath168 with @xmath169 is actually quite ambiguous because of the presence of the undetermined integration constants @xmath115 . due to the different behaviours in uv limit
, we can impose the special choice ( [ eq : spectial choice for const ] ) to ( [ eq:54 ] ) , as a result it yields to a different behavior of @xmath168 with @xmath169 from the qft result in @xcite .
however , we need to emphasize that this comparison with qcd is strictly significant only if the theories with same number of colors and flavors are considered , otherwise theories with different numbers of colors or flavors could have different behaviors . it has turned out that , by using ( [ eq : modified mkk ] ) the string tension is given as , @xmath170.\label{eq:56}\ ] ] imposing the special choice ( [ eq : spectial choice for const ] ) , we have @xmath171 . in this sense , we can naively conclude that the string tension increase by the effect of the dynamical flavors and the presence of the magnetic field .
but our result ( [ eq:56 ] ) seems unrealistic if @xmath172 could be holographically interpreted as some qcd tensions , because intuitively speaking the theory should confine less when more flavors ( or magnetic field ) are added
. however this behavior of the theory should depend on which scheme is chosen and where some observable is kept fixed , since theories with different @xmath2 are actually different as mentioned @xcite .
nevertheless we are not clear about whether the opposite behavior in ( [ eq:56 ] ) corresponds to large @xmath0 limit or the choice ( [ eq : spectial choice for const ] ) for the undetermined integration constants in our theory .
we believe a future study about this is also needed . in ads
/ cft , a baryon is a wrapped d - brane on the extra dimensions @xcite .
accordingly , a baryon vertex is a wrapped d4-brane on @xmath26 in the witten - sakai - sugimoto model . and
it corresponds to the deep ir of the dual field theory since it is localized at the radial position i.e. the holographic direction .
so with the euclidean version of the backreaction solution in the confined case , we can easily read the wrapped d4-brane action , @xmath173 here @xmath174 is the tension of the d4-brane .
using our solution in the confined case at @xmath106 ( i.e. the ir value of the radial direction ) , we have the baryon mass which is given as @xmath175,\ ] ] where @xmath176 for the special choice ( [ eq : spectial choice for const ] ) it gives @xmath177 .
therefore , the baryon mass also increases by the modification of the flavor dynamics and the presence of the magnetic field .
the comments are similar as in the previous subsections . in this subsection
, we test our gravitational solution by evaluating the mass spectrum of the vector mesons .
let us just discuss about the method to obtain the mass spectrum of vector mesons briefly ( see @xcite and @xcite to review the details of this part ) .
as in @xcite , by considering the fluctuation of a gauge field on the d8-brane , we can find the spectrum of the vector mesons with our backreaction solution .
the mass spectrum of the vector mesons is given as , @xmath178\gamma_{n},\ \
n\geq1,\ ] ] and @xmath179 s are given from the following equations @xmath180+h\left(r\right)\gamma_{n}\psi_{n}=0,\label{eq:57}\ ] ] where @xmath181,\nonumber \\
h\left(r\right ) & = \ \frac{9}{4}\frac{e^{-3r/2}}{\left(1-e^{-3r}\right)^{7/6}}\left[1+\epsilon_{f}\left(-3\phi_{1}+4\lambda_{1}+\widetilde{\lambda}_{1}+8\nu_{1}\right)\right].\label{eq:58}\end{aligned}\ ] ] the normalization condition has to be satisfied for @xmath182 , @xmath183 to solve the equation ( [ eq:57 ] ) and ( [ eq:58 ] ) , we could follow the wkb method as in @xcite . since the processes of the calculations based on our backreaction solution is similar as the case without the magnetic field @xcite ,
the only difference is that the meson mass increases and is affected by the magnetic field .
the reason has been commented as before .
so we will not discuss more details and just give the theoretical method we used here .
in this section , we are going to discuss the main subject of this manuscript , i.e. study the thermodynamics and holographic renormalization of this model , by our magnetic - dependent solution . through the holographic formula @xmath184
, the euclidean gravity action is related to the free energy of this model .
as we are going to discuss the thermodynamics of this model , we need to evaluate the euclidean onshell action , taking into account the backreaction by our magnetic - dependent solutions . and the euclidean version of the type ii a supergravity action could be obtained by a wick rotation from ( [ eq : action with dynamic flavors ] ) , which is @xmath185\nonumber \\ & \ \
-\frac{n_{f}t_{8}m_{kk}}{\pi}\int d^{10}x\frac{\sqrt{-\det\left(g+2\pi\alpha^{\prime}f\right)}}{\sqrt{g_{44}}}e^{-\phi}.\label{eq : euclidean action}\end{aligned}\ ] ] however , the onshell action ( [ eq : euclidean action ] ) is divergent if inserting our solutions in confined or deconfined case .
since we would like to compare the free energy of this model with different backreaction solutions , we have to renormalize the theory holographically .
the renormalized gravity action could be written as @xmath186 @xmath187 is the euclidean version of the type ii a supergravity action ( [ eq : euclidean action ] ) and @xmath188 , @xmath189 , @xmath190 is gibbons - hawking ( gh ) term , the bulk counterterm and the d8-brane counterterm respectively . in string frame
, they are given as ) . ]
@xmath191,\label{eq : c.t .
term}\end{aligned}\ ] ] where @xmath192 are two constants for the case of smeared d8-branes and @xmath193 is the determinant of the metric at the uv boundary i.e. the slice of the 10d metric fixed at @xmath194 with @xmath195 .
@xmath196 is the trace of the boundary extrinsic curvature whose explicit form in our notation is @xmath197 and @xmath198 then we are going to evaluate all the terms in ( [ eq : euclidean action ] ) and ( [ eq : c.t .
term ] ) by our magnetic - dependent solutions both in confined and deconfined case . evaluating the action ( [ eq : euclidean action ] ) and ( [ eq : c.t . term ] ) by our magnetic - dependent solution for confined case , we have the following onshell actions ( up to the first order on @xmath76 ) @xmath199,\nonumber \\
s_{gh } & = -a\mathcal{v}\left[-\frac{7}{4}+\frac{19}{6\varepsilon}+\epsilon_{f}h_{2}\right],\nonumber \\ s_{c.t.}^{bulk } & = a\mathcal{v}\left[\frac{5}{3\varepsilon}+\epsilon_{f}h_{3}\right],\nonumber \\ s_{c.t.}^{d8 } & = a\mathcal{v}\epsilon_{f}\left[\frac{2\chi_{1}^{c}-8\chi_{2}^{c}}{3\left(3\right)^{1/6}\varepsilon^{7/6}}+\frac{14\chi_{1}^{c}-8\chi_{2}^{c}}{12\left(3\right)^{1/6}\varepsilon^{1/6}}\right].\label{eq : onshell confined}\end{aligned}\ ] ] where @xmath200 in order to cancel all the divergences in ( [ eq : onshell confined ] ) and ( [ eq : confined onshell - 2 ] ) , we have to choose @xmath201 as we can see from ( [ eq : confined onshell - 2 ] ) , there are magnetic - dependent divergences . however , for a constant magnetic field , it is possible to choose the @xmath116-dependent constants ( [ eq : counter term constant confined ] ) in the counterterm action @xcite .
therefore the renormalized action for the backreaction case reads @xmath202\right).\label{eq : confined ren}\ ] ] as in the confined case , the euclidean version of the onshell action ( plus the gh term ) ( which is the gibbs free energy ) is also divergent , thus it must be renormalized by approaching the counterterms in ( [ eq : ren action ] ) .
the functional form of each term in ( [ eq : ren action ] ) takes the same formulas as ( [ eq : euclidean action ] ) and ( [ eq : c.t . term ] ) respectively , however it needs to be evaluated by our deconfined solution .
therefore we have , @xmath203,\nonumber \\ s_{gh } & = -a_{t}\mathcal{v}\left[-\frac{7}{4}+\frac{19}{6\varepsilon}+\epsilon_{ft}g_{2}\right],\nonumber \\ s_{c.t.}^{bulk } & = a_{t}\mathcal{v}\left[\frac{5}{3\varepsilon}+\epsilon_{ft}g_{3}\right],\nonumber \\ s_{c.t.}^{d8 } & = a_{t}\mathcal{v}\epsilon_{ft}\left[\frac{2\chi_{1}^{d}-8\chi_{1}^{d}}{3\left(3\right)^{1/6}\varepsilon^{7/6}}+\frac{\chi_{1}^{d}-16\chi_{1}^{d}}{6\left(3\right)^{1/6}\varepsilon^{1/6}}\right].\label{eq : deconfined onshell - 1}\end{aligned}\ ] ] where @xmath204 as we can see , the `` bulk counterterm '' @xmath205 cancels the @xmath206 divergences only as the confined case .
we thus have introduced the additional counterterm , i.e. the `` flavor counterterm '' @xmath207 which is related to the d8-branes , to cancel the remaining divergences , the value of the same combination on some background as being a reference . ] .
consequently , we have to choose the following values ) are enough to cancel all the divergences if the @xmath116-dependent constants in the counterterms are allowed .
since there are not charge - dependent divergences in the calculations of @xcite in the deconfined case , the divergent terms in our calculations ( [ eq : deconfined onshell - 1 ] ) should definitely return to @xcite by setting @xmath208 .
nevertheless , we find the divergent terms of @xmath209 is a bit different and the values of @xmath210 and @xmath211 can never return to the choice in @xcite even for setting @xmath208 .
we have checked this problem and found that , according to our solution by setting @xmath208 , all the calculations would be exactly same as @xcite if evaluating the minkowskian action instead of the euclidean action .
consequently , the resultant calculations are a little different from @xcite if evaluating the euclidean action by our solution , even for setting @xmath208 . ]
@xmath212 to cancel all the divergences in ( [ eq : deconfined onshell - 1 ] ) and ( [ eq : deconfined onshell - 2 ] ) . with these choices , we have the renormalized action in the deconfined case which is , @xmath213.\label{eq : deconfined ren}\ ] ]
in this section , let us discuss the phase diagram of this holographic model in the presence of a magnetic field and compare the diagram with lattice qcd .
since there is no chemical potential through our setup , we will thus focus on the case of finite temperature and zero chemical potential in qcd . since our goal is to quantify the effects from the flavors on the critical temperature in the presence of the magnetic field when the phase transition happens between the confinement and deconfinement phase .
therefore , we just need to compare the free energy from the renormalized onshell action . and we should first calculate the pressure @xmath214 both for confined and deconfined phase by using , @xmath215 since we have introduced the additional boundary terms for the flavor branes in ( [ eq : c.t . term ] ) , it admits the holographically renormalized bulk action .
moreover , returning to the case of the probe limit is also allowed definitely from our backreaction case .
hence in the probe approximation , we have the onshell d8-brane action with the u - shape embedding i.e. @xmath125 , which is @xmath216 where , @xmath217 here we have expanded the action in small @xmath116 limit .
it can be found there are two divergent terms in the onshell action ( [ eq:70 ] ) in the uv .
therefore , in order to cancel the divergences in ( [ eq:70 ] ) , we have to choose @xmath218 in the d8-brane counterterm ( [ eq : onshell confined ] ) for the probe approximation . in a word , we obtain the renormalized action ( bulk plus flavor brane ) in the probe approximation as , @xmath219.\ ] ] accordingly , we have the pressure in the probe approximation for the confined case , @xmath220.\label{eq:74}\ ] ] for the deconfined case ( similarly as in the confined phase ) , we also have the onshell d8-brane action with the parallel embedding , which is @xmath221 where @xmath222
so we need the following choice , @xmath223 for the additional flavor brane counterterm in ( [ eq : deconfined onshell - 1 ] ) .
obviously , in the probe approximation the renormalized onshell d8-brane action reads @xmath224.\ ] ] and its pressure is , @xmath225.\label{eq:79}\ ] ] consequently , we can obtain the phase diagram in the probe approximation by comparing the pressure ( [ eq:74 ] ) and ( [ eq:79 ] ) with the equation @xmath226 since the contribution form @xmath227 in @xmath77 could be neglected . ]
, it gives @xmath228-\frac{81}{16\pi}\frac{n_{f}}{n_{c}}\frac{b^{2}}{m_{kk}^{4}}\left[1-\frac{2\pi^{3/2}}{3\gamma\left(\frac{1}{6}\right)\gamma\left(\frac{4}{3}\right)}\right]\label{eq:76}\ ] ] where @xmath229 so from ( [ eq:76 ] ) and ( [ eq:77 ] ) , we could conclude that , at zero chemical potential without the magnetic field , the critical temperature increases by the effect of the flavors ) is quantitative same as @xcite definitely . ]
( see also @xcite ) .
and we also notice that the contribution from the magnetic field is quadratic for any @xmath2 .
moreover , ( [ eq:76 ] ) shows @xmath9 decreases when @xmath169 increases ( as shown in figure 2 ) which is in agreement with the lattice qcd results @xcite . ) or ( [ eq:69 ] ) . *
lower * : the phase diagram in t - b plan from some lattice qcd results in @xcite .
@xmath230 is a parameter in the lattice calculations . ] ) or ( [ eq:69 ] ) .
* lower * : the phase diagram in t - b plan from some lattice qcd results in @xcite .
@xmath230 is a parameter in the lattice calculations . ]
let us turn to the case of backreaction . to get the phase diagram , first we need to imposing ( [ eq : modified mkk ] ) and (
[ eq : modified t deconfined ] ) on ( [ eq : confined ren ] ) and ( [ eq : deconfined ren ] ) , thus obtain the pressure of each phase as , @xmath231,\nonumber \\
p_{\mathrm{deconf}. } & = \frac{64\pi^{4}n_{c}^{2}t^{6}\lambda_{4}}{2187m_{kk}^{2}}\left[1+\frac{\lambda_{4}^{2}}{12\pi^{3}}\frac{n_{f}}{n_{c}}\left(-\frac{4}{7}-a_{1}-a_{2}+5a_{3}+4b_{3}+\frac{377}{388}q_{b}^{2}\right)\right].\label{eq : pressure smeared}\end{aligned}\ ] ] in order to obtain the critical temperature at the phase transition point , we could solve the equation @xmath232 as in the probe approximation .
then we find the following relation between the critical temperature @xmath9 and the magnetic field @xmath169 , @xmath233 where @xmath234 and we have also imposed the spectial choice ( [ eq : constants - 2 confined ] ) ( [ eq : spectial choice for const ] ) ( [ eq : constant for deconfined phase - 2 ] ) ( [ eq:47 ] ) to ( [ eq : smeaed phase diagram - 1 ] ) .
so similar to the case of the probe approximation , ( [ eq : smeaed phase diagram - 1 ] ) shows the critical temperature increases by the effect of the flavors without magnetic field .
but notice that the factor in front of the magnetic field is positive , it shows that the behavior of ( [ eq : smeaed phase diagram - 1 ] ) is totally different from the probe approximation and is not in agreement with the lattice qcd results @xcite .
in fact ( [ eq : smeaed phase diagram - 1 ] ) should be a scheme - dependent statement thus it depends on the choices of the appropriate interpretation ( also the numbers of colors and flavors ) . as mentioned ,
since we are less clear about the full relations between the integration constants and the constant magnetic field @xmath169 , we therefore use the same ansatz as the most simple choice as @xcite for the undetermined integration constants .
however the behavior of @xmath9 with @xmath169 ( [ eq : smeaed phase diagram - 1 ] ) is actually quite sensitive to the relations between the integration constants and the constant magnetic field . on the other hand ,
we have also mentioned the deconfined geometry does not strictly correspond to the deconfinement phase in the dual field theory .
accordingly it is not surprised that ( [ eq : smeaed phase diagram - 1 ] ) does not agree with lattice qcd , but the less clear relation of the integration constants leads to an ambiguous behavior of @xmath9 with @xmath169 if imposing ( [ eq : constants - 2 confined ] ) ( [ eq : spectial choice for const ] ) ( [ eq : constant for deconfined phase - 2 ] ) directly . hence in order to return to the probe approximation , we have to additionally require , @xmath235 and @xmath236 although we do not obtain the explicit relations between all the integration constants and the constant magnetic field .
otherwise , the original geometrical configuration might be totally deformed and the perturbative solution both for confined and deconfined phase would become unreliable . besides , there may be an alternative comparison scheme available , which is motivated by the in heavy - ion collision experiment phenomenally . before the collision ,
the nuclei are confined and an extremely external ( strong ) magnetic field is created when the collision happens .
accordingly , we treat the confined case without the magnetic field as the initial state of the nuclei , then the magnetic field @xmath169 induces the phase transition when the collision happens .
we therefore need to solve @xmath237 to obtain the phase diagram , it would be more simple , @xmath238-\frac{n_{f}}{n_{c}}\frac{1131b^{2}}{448\pi m_{kk}^{4}}.\label{eq:69}\ ] ] ( [ eq:69 ] ) is a desired phase diagram since it qualitatively agrees with the lattice results @xcite as well . in this scheme , although the explicit relation between the integration constants and the magnetic field is not explicitly clear , the behavior of the diagram would be unique .
this is due to that the contribution from the magnetic field to the pressure of the deconfined phase is always positive ( see also @xcite ) . and on the other hand , the probe limit of this model is definitely allowed to return .
therefore it gives the unique behavior as ( [ eq:69 ] ) .
however we must keep in mind that the scheme of the interpretation we choose .
in this paper , we have constructed gravitational solutions as a magnetic - dependently holographic background in the witten - sakai - sugimoto model , by considering the backreaction of the flavors and the magnetic field .
thus it corresponds to a quantum field theory ( or qcd ) with dynamical flavors in an external magnetic field .
we have proved out our gravitational solutions satisfy their equations of motion explicitly in the first order of @xmath1 . and
the solutions are analytic both in confined and deconfined case at low ( zero ) or high ( finite ) temperature . therefore these solutions are able to study the the influence of dynamical flavors in an external magnetic field as a holographic version of @xcite . in order to determine the integration constants in our solutions
, we require the backgrounds are completely regular in the ir region of the dual field theory as the unflavored case since the flavors are small perturbations . on the other hand
, we also try to turn off the sources or vevs of some gauge invariant operators in the dual field theory as another constraint
. however the calculation shows it is not enough to determine all the integration constants just by these two constraints .
so we should keep those undetermined integration constants as some generic parameters in our theory . in order to compare our magnetic - dependent case with @xcite
, we naively chose the same value for the undetermined integration constants as @xcite , to study some physical properties about hadronic physics in an external magnetic field , such as the running coupling , string tension , baryon mass , vector meson mass spectrum .
we find the uv behavior of the running coupling is not affected by the presence of the magnetic field . and the string tension , the mass of baryon or meson increase by the flavor or the magnetic field .
but we need to keep in mind these behavior should depend on which scheme is chosen and where some observables is kept fixed in the theory , since theories with different numbers of flavors are different . additionally , because of the simply choice as @xcite for those undetermined integration constants , the results in this part are not strictly rigorous thus some of them might still seem unrealistic .
moreover , it shows the physical significance of our work by studying the holographic renormalization and thermodynamics with our magnetic - dependently gravitational solution .
we employ the counterterm @xcite and its covariant formula @xcite for this model then evaluate it by our magnetic solution .
the motivation for studying this counterterm is to renormalize the free energy , to study the hawking - page transition holographically in the presence of the magnetic field . in some applications of the witten - sakai - sugimoto model ,
holographic renormalization may not be necessary for studying the phase transition .
since those concerns are the difference of the free energy of the various configurations of the flavor branes in the same background , which is not the hawking - page transition of this model .
so the difference of the free energy could be finite in those approaches ( such as @xcite ) .
however , in our calculations , holographic renormalization is needed since we , more than that , also consider the transition between differently geometric background . and according to our calculations , if the parameters in the covariant counterterms are allowed to depend on the magnetic field as @xcite , we find the present counterterms are enough to cancel all the divergences . in particular , after the holographic renormalization , we have concentrated the attention on the holographic phase diagrams in the presence of the magnetic field , and compare it with lattice qcd results . in our backreaction case , we find the pressure of both phases evaluated by our magnetic - dependent solution agrees with @xcite qualitatively although we have kept the undetermined integration constants generic as mentioned .
while the behavior of the phase diagram is a bit ambiguous in the backreaction case , in the probe approximation it is clear and in agreement with the lattice qcd @xcite qualitatively ( figure 2 ) .
thus it could be interpreted as the inhibition of confinement or chirally broken symmetry by the magnetic field holographically .
finally , let us comment something more about our work .
as an improvement to @xcite , we have employed the technique used in @xcite to take into account the backreaction from flavors and the magnetic field .
because of the presence of the magnetic field , actually we need to solve a set of highly non - linear equations of motion first to obtain a magnetic - dependently gravitational solution , as shown in ( [ eq : eoms in confined phase ] ) and ( [ eq : eom for deconfined phase ] ) . since it is hopeless to find an analytic solution from these extremely complicated equations , we solve them by keeping the leading @xmath239 terms . while it is a challenge to keep all the orders of the dbi action to solve analytically , some numerical calculations might be worthy . besides , during our calculations
, we have restricted that @xmath5-branes are placed at antipodal points of @xmath10 - circle in the confined phase .
so to extend this part to the non - antipodal case would be natural , and the chiral symmetry could also be restored after deconfinement transition .
moreover , it is also interesting to turn on a chemical potential and a magnetic field together on the flavor branes in this framework , since a similar phenomena , named as `` inverse magnetic catalysis '' , has also been found by using this model in the probe approach of @xcite .
however , there would be a non - vanished chern - simons term necessarily ( @xmath71 gauge field strengh ) .
there have been some discussions about the inverse magnetic catalysis in the einstein - maxwell system as @xcite .
however , as a difference from einstein - maxwell system and also a computational challenge , we have to consider the additional romand - romand field in the bulk and the non - vanished chern - simons ( or wess - zumino ) term if taking into account the backreaction from the flavor branes ( full action ) . while the computation is difficult , it would be quite interesting for a future study . ] if turning on the chemical potential and the magnetic field together as @xcite .
accordingly , it would be more difficult to search for an analytic solution even in the expansion of small baryon charge , magnetic field and @xmath1 in that case since the equations of motion would be complicatedly coupled to each other once the backreaction is considered
. we would like to leave these interesting topics for a future study to improve our calculations about holographic qcd .
this work is inspired by our previous work @xcite in ustc , and also by @xcite the recent research on the magnetic field in heavy - ion collision from our colleagues .
and we would like to thank andreas schmitt , prof .
qun wang and dr .
chao wu for helpful discussions .
f. bigazzi , a. l. cotrone , l. martucci , l. a. pando zayas , `` wilson loop , regge trajectory and hadron masses in a yang - mills theory from semiclassical strings , '' phys . rev .
d 71 , 066002 ( 2005 ) [ hep - th/0409205 ] . | we construct the gravitational solution of the witten - sakai - sugimoto model by introducing a magnetic field on the flavor brane . with taking into account their backreaction ,
we re - solve the type iia supergravity in the presence of the magnetic field .
our calculation shows the gravitational solutions are magnetic - dependent and analytic both in the confined and deconfined case .
we study the dual field theory at the leading order in the ratio of the number of flavors and colors , also in the veneziano limit . some physical properties related to the hadronic physics in an external magnetic field
are discussed by using our confined backreaction solution holographically .
we also study the thermodynamics and holographic renormalization of this model in both phases by our magnetic - dependent solution .
since the backreaction of the magnetic field is considered in our gravitational solution , it allows us to study the hawking - page transition with flavors and colors of this model in the presence of the magnetic field .
finally we therefore obtain the holographic phase diagram with the contributions from the flavors and the magnetic field .
our holographic phase diagram is in agreement with lattice qcd result qualitatively , which thus can be interpreted as the inhibition of confinement or chirally broken symmetry by the magnetic field .
si - wen li and tuo jia _ department of modern physics , _ + _ university of science and technology of china , _ + _ hefei 230026 , anhui , china _ |
presently there is a great deal of interest in properties of amphiphilic copolymers because of the variety and extent of conformational transitions they show under a temperature or ph variation @xcite .
for instance , as such copolymers contain both hydrophobic and hydrophilic ( polar ) monomers they can form monomolecular micelles with a hydrophobic core in water , as well as inverted micelles in apolar solvents . furthermore , these systems have an interesting analogy to globular proteins in the folded state @xcite . for a discussion of similarities between `` coarse grained '' features of proteins and those of amphiphilic polymers we refer to ref .
@xcite .
linear copolymers , in particular linear diblock copolymers , would aggregate under certain conditions , giving rise to a fascinating range of self assembled supramolecular ordering in the absence of solvent @xcite , and therefore the main issues of study traditionally have been the critical micelle concentration and the solid state morphology . on the other hand ,
the possibility of forming monomolecular micelles is greatly increased by using more complicated topologies , for instance star or comb polymers @xcite .
random branching of polymers is indeed quite common , but carefully controlled polymerisation procedure allows one to obtain regular stars , which present a particularly nice object for theoretical study due to their symmetry and simplicity .
star polymers are believed to find a number of future applications for coating , as additives and possibly in drug delivery systems related to their low viscosity and other interesting properties .
the aim of the present paper is to investigate the behaviour of diblock star copolymers comprising both hydrophilic and hydrophobic monomers within the same arm in dilute solution .
we shall not consider here the case of miktoarm star copolymers comprising chemically homogeneous arms made of two or more different monomers @xcite .
our study will be carried out by means of both the analytical gaussian self consistent theory and monte carlo simulation in continuous space
. such combination of techniques would permit us to assess the validity of the analytical theory , which up to now was independently developed and used by our two groups in milan and dublin to investigate the behaviour of random linear copolymers , though in seemingly distinct formulations . in appendix of this article we shall demonstrate the equivalence of these two versions of the theory by relating the free energy expressions in both .
we would like to emphasise that this general analytical approach is the only one , to the best of our knowledge , which can explicitly account for the details of intramolecular conformation , thus providing more refined characteristics of the polymer than simply the overall molecular size or density profile .
the predicted intramolecular conformations may then be compared with those from direct computer simulations .
computer simulations of star polymers ( see e.g. refs .
@xcite for recent reviews ) have been carried out by a number of techniques , ranging from monte carlo methods , both on and off lattice , to brownian dynamics and molecular dynamics @xcite . most of works on block copolymer stars , however , dealt with the molten state and the supramolecular organisation under such conditions .
the only study on diblock star copolymers in dilute solutions similar to ours , as far as we are aware , was carried out by means of a lattice monte carlo simulation in ref .
the latter paper appears to contain a somewhat confusing description of the adopted interactions model and its status .
judging by the actually observed star conformations , those correspond to the case in which both types of units are strongly immiscible with each other and also both tend to avoid the solvent molecules .
this is not , of course , what one would really call an amphiphilic polymer , in which one type of units should `` like '' solvent while the other should not .
besides , there are quite obvious limitations of the lattice model as such ( see e.g. discussion in ref .
@xcite ) , which become even more serious for a star polymer . indeed
, the core monomer on a lattice has a great difficulty to move at all due to its high coordination number ( equal to the number of arms @xmath1 ) .
the resulting rejection of all attempted moves for some kind of motions is called a quasi
nonergodicity , a problem which is hard to deal with even for linear copolymers .
if such a situation occurs , it practically means that the results of the simulation may no longer be trusted .
a number of complicated non local moves involving the core monomer have been suggested to alleviate this problem @xcite .
however , such moves are not permitted in case of dynamic monte carlo needed for study of , for instance , metastable states . moreover , for polymers preserving the integrity of links ( i.e. if chain segments can not pass through each other ) mixing of local and non local moves may result in subtle topological effects leading to a non uniform sampling of the phase space and eventually to breaking of the balance condition needed for convergence of monte carlo scheme to the state of equilibrium @xcite . for these and other reasons
it is important to examine the conformations of diblock star copolymers in continuous space based on the correct model of hydrophobic hydrophilic interactions .
the plan of our paper is as follows .
first , we briefly summarise the analytical approach used .
then , we present the corresponding results by discussing both the molecular size and the intramolecular conformation .
next , we introduce the model for monte carlo simulation and present the results from such an approach performing their comparison with those of the analytical theory . finally , we discuss the validity and limitations of the gaussian approach .
we consider regular star copolymers formed from @xmath1 equal arms , each consisting of @xmath2 monomers , with a total of @xmath3 monomers , including also the core monomer . on each arm ,
the monomer beads are sequentially numbered from @xmath4 to @xmath2 starting from the branch point , which is labelled as @xmath5 .
the beads are connected by freely
jointed bond vectors of unit length oriented outwards .
furthermore , each arm is formed from two blocks of equal lengths , each comprising @xmath6 beads .
one block consists of hydrophobic beads ( or , in brief , h - beads , thus forming an h - block ) , while the other block consists of hydrophilic , or polar , beads ( p - beads and p - block ) . a star polymer with the h - block topologically adjacent to the branch point ( and thus with an outer p - block ) will be called an inner - h star , while a star polymer with the opposite connectivity will be called an outer - h star .
obviously , the topological location of the h - beads is quite important because in water they would experience a mutually attractive interaction , which eventually determines the chain conformation , and in particular the formation of globular clusters . for simplicity
, we shall assume that the type of the bead at the branch point is the same as of the beads in the inner blocks .
the equilibrium conformation of the polymer is then determined by a self consistent free energy minimisation based on the procedure employed in refs .
@xcite for linear random copolymers .
the intramolecular free energy ( see appendix for more details ) consists of the configurational entropy and the intramolecular interactions terms .
the latter ones include : ( i ) the two body interactions , which can be attractive , repulsive or even vanishing depending on the type of beads involved ; ( ii ) the repulsive screened interactions accounting for the effective `` thickness '' of the beads ; ( iii ) the repulsive three body interactions .
the last two contributions are assumed to be independent from the type of the interacting beads , unlike the first contribution .
following refs .
@xcite , we may write it in the gaussian approximation , as a sum over all pairs of beads of the probability density of contacts multiplied by the interaction constants specifying the sign and strength of these interactions .
thus , the free energy contribution due to the two - body interactions in @xmath7 units may be written as , @xmath8 where @xmath9 is the mean squared distance between beads @xmath10 , and @xmath11 are the interaction constants .
the positive and negative signs of these constants mean repulsive and attractive interactions respectively .
furthermore , following refs .
@xcite we shall assume in addition that , @xmath12 where @xmath13 indicates a p - bead and @xmath14 an h - bead .
a rationale for the latter equation was recently presented by two of us ( see eqs .
( 8 - 10 ) in ref .
this particular parametrisation of @xmath11 constants results after the effective integration over the solvent molecules degrees of freedom proceeding from the hamiltonian with a contact monomer
solvent molecules interaction upon the solution incompressibility condition . for convenience
, we may ascribe to each bead a value @xmath15 , specifying its p- or h - character , thereby writing @xmath16 in the form , @xmath17 thus , @xmath18 may be taken as a measure of the copolymer s amphiphilicity , while @xmath19 characterises the strength of the h - p two body interactions . in the present paper , we take @xmath20 , so that the latter interactions vanish .
it should be remembered , however , that these beads still experience the weaker repulsive interactions ( screened and three body ) . following the general procedure outlined in refs .
@xcite within the framework of gaussian approximation we can write the intramolecular free energy as a unique function of the average scalar products among the bond vectors , @xmath21 , which are thus the variables subject to optimisation
. the only difference of a star polymer with a linear chain is then , clearly , only in the expression yielding the mean squared distances between beads ( on the same or on different arms ) in terms of this set of the chosen variational variables .
this expression is given and discussed in detail in ref .
@xcite and is not reported here for brevity .
calculations were carried out for 3-arm star copolymers with 16 beads per arm and for 4-arm star copolymers with either 12 or 16 beards per arm . in this way , we could compare stars with either the same number of beads ( i.e. , @xmath22 beads ) or with the same arm length ( i.e. , @xmath23 beads ) .
we chose not to impose any symmetry properties upon statistical averages over the arms .
this was done to make sure that no asymmetrical conformations were lost ( see discussions in ref .
@xcite ) , although the problem becomes more expensive in terms of computations .
a description of the numerical free energy minimisation procedure and the details of all equations expressing various quantities of interest for describing the conformation of a star may also be found in ref .
@xcite . upon increasing the degree of amphiphilicity @xmath18 both inner - h and outer - h stars show clustering of the h - beads into dense globules , but with an essential difference .
formation of a single globule is easily achieved for an inner - h star through a continuous ( second order ) transition due to the initial close vicinity of the mutually attractive h - beads thanks to the star connectivity . on the contrary , for an outer - h star
this is much more difficult to achieve , so that formation of a single globule proceeds through a discontinuous ( first order ) transition .
furthermore , for these stars a long lived metastable state consisting of @xmath1 small clusters at the arm ends is also possible .
the smooth collapse transition of an inner - h star can be followed through the radii of gyration dependence on @xmath18 , as shown in fig .
[ fig : fa ] for the 3-arm star ( results for the 4-arm stars show a similar pattern and are not shown for brevity ) .
it is useful to consider the mean squared distance of the h - beads from their own centre of mass , @xmath24 , which decreases monotonously upon increasing @xmath18 ( i.e. increasing attractive interactions ) . at the same time , @xmath25 ,
analogously defined for the p - beads , as well as @xmath26 , defined as the mean squared distance of all beads from the common centre of mass , show an initial decrease , which then turns into a steady increase with increasing @xmath18 .
this behaviour may be easily explained by the increased repulsive potential among the p - beads .
it should also be noted that this trend is in addition enhanced by the pull exerted by an increasingly tight core , which brings the arms closer together . as a result , the external p - blocks are highly stretched outwards .
an inner - h star forms a tight globular core easily if the free energy minimisation uses a random walk conformation as the starting point . in the case of an outer - h star the same procedure , however ,
only leads to a metastable state at large values of @xmath18 . here , the h - beads form @xmath1 small well separated subglobules at the end of each arm .
coalescence of these subglobules is prevented by their large spatial separation imposed by the inner p - blocks , which tend to be as far as possible from each other due to their mutual repulsion .
thus , as shown in fig .
[ fig : fb ] with dotted curves for the 3-arm star , in the metastable state @xmath25 increases monotonously with increasing @xmath18 , while @xmath26 and @xmath24 show a shallow dip when the h - blocks do collapse . further increase of @xmath24 at larger @xmath18 is related to the fact that , while the individual clusters are actually shrinking , they are being pushed afar from the center of mass of the h - beads by the simultaneous stretching of the p - blocks .
the stable equilibrium state of an outer - h star can only be found when using a fully collapsed h - star homopolymer with the same total number of beads as the starting point for free energy minimisation . in this stable state ,
all h - beads cluster together in a single globule .
this globule also contains the p - bead of the branch point .
the latter feature , however , is not supported by the simulation results in this paper , showing that in the true stable state the core p - bead would actually escape from the hydrophobic core . apart from this arguable behaviour of the branch point , all other p - beads of the topologically inner blocks form closed swollen loops with an overall daisy like pattern .
these loops tend to be as far as possible from the globular core ( as in the stable state of an inner - h star ) , but are seriously constrained by the arm connectivity . as a result , in the stable globular state the overall molecular size of an outer - h star is quite smaller than the size of the respective inner - h star . on the other hand ,
it is interesting to note that the size of the dense core of h - beads depends only on the total number of beads , but not on the type of star and arms number , i.e. the same @xmath24 values are obtained for inner - h and outer - h stars with @xmath27 or @xmath28 arms at fixed values of @xmath18 and @xmath29 .
this indicates that the molecular topology becomes irrelevant inside dense globules and that their size is dictated only by the space filling requirement , in agreement with previous results @xcite .
conversely , for an inner - h star , using the globular conformation of a fully collapsed h - star homopolymer as the starting point of free energy minimisation results in yet another metastable state . in such a state , the globular core formed by h - beads has essentially the same size as in the stable state , though being less ordered . however , the p - beads tend to be much closer to the common centre of mass , though still protruding outside the globule .
this conformation was quite unexpected due to the strong repulsive interactions experienced by the p - beads .
as a possible rationalisation for this , we speculate that such a metastable state encompasses a certain number of `` knotted '' arms with p - blocks being trapped for entropic reasons within the h - core .
we do not report such results in fig .
[ fig : fa ] since they are not supported by the computer simulation results .
one should realise that the gaussian variational theory is , in fact , a mean field theory which may contain some `` unimportant '' local minima on the free energy surface . these may either be unstable in some directions or separated by too shallow a barrier from the main minimum , which may thus could be easily overcome due to thermal fluctuations .
the latter point is consistent with a relatively poor convergence encountered during the numerical free energy minimisation for this particular metastable state : the final mean squared gradient of the free energy turned out to be somewhat larger than expected and all further attempts to improve the numerical procedure were plagued by numerical instabilities . in order to discuss the intramolecular conformation ,
we shall report some representative results obtained for the 3-arm stars with @xmath18=1.0 . at large values of @xmath18 for an inner - h star the h - beads of the globular core experience a strong attractive interaction , which leads to the formation of a random globule and further to an ordered globule , while the p - blocks stretch outside because of their repulsion .
this intramolecular conformation is best described via the mean squared distances of the beads belonging to a given arm measured from the centre of mass of the polymer .
this quantity is shown in fig .
[ fig : fc]a as a function of the position along the arm .
we show the plot for only one arm because of the statistical symmetry among the arms found in all cases .
the ordering of the hydrophobic core is indicated by the zigzag pattern of the h - beads ( black circles in fig .
[ fig : fc]a ) , which means that the beads lie alternatively on two concentric shells centred at the centre of mass .
correspondingly , the scalar products among the bond vectors connecting the h - beads , @xmath30 at fixed @xmath31 tend to be equal to @xmath32 , as was previously reported for short homopolymer chains @xcite .
also , the significantly different distances from the centre of mass for the h- and p - beads give rise to rather different shapes of the density profiles .
these density profiles are shown in fig .
[ fig : fd ] using the following dimensionless variables , @xmath33 where @xmath34 stands for any of the three functions @xmath35 , @xmath36 or @xmath37 .
note that each curve is normalized to the fraction of the corresponding beads within the molecule , i.e. @xmath38 , index @xmath39 being @xmath40 , @xmath41 , or @xmath42 .
the main features here are the sharp peak for the h - beads related to their tight clustering in the core , and the very broad profile for the p - beads related to their broad distance distribution away from the centre of mass . as a result
, the overall profile of @xmath37 is strongly skewed at large values of @xmath43 . however ,
because of the molecular connectivity , no more complicated pattern , such as say a two peak profile , is observed . in the case of the 3-arm outer - h star , the daisy - like pattern of the conformation in the stable state is clearly seen via the mean - squared distances of the beads of a given arm from the centre of mass reported in fig .
[ fig : fc]b ( solid line ) .
in particular , the p - beads ( white circles ) tend to be as far away from the center of mass as possible , thereby maximising their separation , but being constrained by the molecular connectivity they return back to the core of the h - beads .
we should note , however , that according to the present results the polar branch point ( labelled as @xmath44 ) remains trapped within the hydrophobic core .
this pecularity though is not supported by computer simulations results as we have already indicated .
the density profiles for this polymer in fig .
[ fig : fe ] reflect similar features .
the density of the h - beads is well localised with a sharp peak at around @xmath45 corresponding to a tight packing of these beads in the core . on the contrary ,
the density of the p - beads is rather delocalised with a shallow peak around @xmath46 corresponding to extended loops of the p - beads .
there is also a slightly non monotonic behaviour observed in @xmath47 at small @xmath48 related to the trapping of the branch point p - bead within the hydrophobic core . as for the total density ,
its profile suggests that overall the polymer conformation is only dense around the centre of mass with a fairly long trailing density tail .
now , if we turn our attention to the conformational structure of the metastable state of the 3-arm outer - h star presented in fig .
[ fig : fc]b ( dashed curve ) , we can clearly see an almost linear increase of @xmath49 with arm index for the p - beads ( white circles ) , reflecting long and rather stretched p - blocks .
however , as the arm index reaches @xmath50 the mean squared distances no longer increase , but remain at the same value with small zig
zag oscillations corresponding to ordered small subglobules of the h - beads formed at the ends of the outstretched arms . as in the case of the stable globular core
discussed before , the size of these subglobules at a given @xmath18 depends only on the number of h - beads they comprise , but not on the number of arms , thus again being dictated by space filling requirement .
accordingly , the density profiles for this metastable state in fig . [ fig : ff ] show that the peak in @xmath47 occurs at smaller @xmath48 than in @xmath51 and it is marginally higher , so that the h - beads are indeed concentrated at the ends of the p - blocks in the form of @xmath1 clusters .
finally , the total density has a shape similar to both of the partial densities , though it is about twice higher and has the width of @xmath51 .
many previous simulations of star polymers were performed on a lattice . however , any lattice model has a number of serious disadvantages and artefacts .
these include rotational anisotropy , tendency for condensed phases to form crystalline structures due to discrete lattice spacing , and dramatic reduction of the acceptance ratio for moves of the core monomer .
thus , we have carried out our simulation in continuous space , which provides by far a more realistic , though somewhat more computationally expensive , description .
the model is implemented for a single star polymer consisting of @xmath52 monomers connected by harmonic springs between any two monomers @xmath31 and @xmath53 within arms of the star , which we denote as @xmath54 .
in addition , all monomers interact with each other via a pair wise short ranged spherically symmetric potential , so that the hamiltonian is @xmath55 this bead and spring model is much more tractable for a monte carlo simulation than the freely jointed model adopted in the analytical part of this work . in any case , since we are interested in rather generic conformational properties of star copolymers , the differences between these implementations of the molecular connectivity should be quite insignificant .
unlike the gaussian theory , where for technical reasons one has to introduce a virial type expansion representing this pair wise potential , here we can use a more realistic two body interaction potential explicitly , @xmath56 thus , monomers are represented by hard spheres of the diameter @xmath57 , with a short ranged lennard
jones attraction of characteristic strengths @xmath58 .
the latter coefficients have the expression akin to eq .
( [ 2 ] ) , @xmath59 where , however , to ensure agreement with eq .
( [ 3 ] ) we should take the _ ansatz _
, @xmath60 so that @xmath61 for any p
h , and h
h pair of monomers respectively .
thus , although the models employed in this section and in the previous one are ideologically the same , they do differ somewhat technically . during simulation
we change the strength of the degree of amphiphilicity @xmath18 , which can be viewed as basically the `` inverse temperature '' , rather than changing the temperature @xmath62 itself . at @xmath63
there is only a hard core repulsion , which is represented by a positive three body coefficient in the virial expansion of the previous section . as a maximal value of the parameter @xmath18 we chose @xmath64 , because in this case the interactions of h
p units are close to those near the theta point corresponding to vanishing of the second virial coefficient ( this is also consistent with the choice of @xmath20 in eq .
( [ 3 ] ) ) .
the monte carlo updates scheme is based on the metropolis algorithm with local monomer moves employed by two of us in ref .
the new coordinate of a monomer can be sought as , @xmath65 , where @xmath66 stands for @xmath67 , @xmath68 and @xmath69 spatial projections and @xmath70 is a random number uniformly distributed in the interval @xmath71 $ ] .
here @xmath72 is some additional parameter of the monte carlo scheme , which , in a sense , characterises the timescale involved in the monte carlo sweep ( mcs ) , the latter being defined as @xmath29 attempted monte carlo steps . in the current model we work in the system of units such that @xmath73 , @xmath74 , which introduces respective reduced variables for length and energy , and in addition
we choose the hard core diameter @xmath75 .
simulations have been carried out for diblock star copolymers with @xmath76 arms and arm length @xmath77 in all cases .
an initial canonical conformation , prepared as a perfect geometrical star in a 2-d plane , is first equilibrated by applying a large number of monte carlo sweeps ( typically well over @xmath78 sweeps ) at @xmath63 .
this procedure allows one to prepare a statistical ensemble ( in our case of @xmath79 ) statistically independent homopolymer stars in the swollen coil state .
even though we formally preserve notations of h- and p - beads , in this case all beads experience hard core repulsions .
this initial statistical ensemble is then used for collapsing each of these stars independently by increasing parameter @xmath18 , so that the h - beads become increasingly hydrophobic while the p - beads remain polar , this being done in two possible ways .
first , one could perform a quasistatic process , in which @xmath18 is changed in @xmath80 small steps until we reach its maximal possible value @xmath81 , allowing a long equilibration of the star after each parameter change . in this way we obtain equilibrium conformation for each of these intermediate values of @xmath18 .
second , we could perform a sudden change @xmath82 and follow the kinetics of the transition of the initial coil conformation to the final collapsed equilibrium state at @xmath83 .
such kinetic procedure is useful for being able to study formation of possible metastable states , which have been predicted by the analytical theory above . as for calculation of statistical averages of various observables of interest , for the kinetics one
is only allowed to average over values in each of the @xmath84 independent members of the statistical ensemble at the same moment in time @xmath85 , whereas , thanks to the ergodicity property , for the equilibrium states one could average over both the ensemble @xmath84 and different moments in time ( i.e. different monte carlo sweeps well separated from each other to improve statistical independence ) .
the latter allows one to obtain much better statistics and hence smoother curves in case of equilibrium measurements , while getting the same quality for the kinetics would seem too time consuming .
now , we are going to compare the results from the computer simulation with those of the analytical sec . [
sec : analres ] .
a typical snapshot of the equilibrium collapsed state of the inner - h star is shown in fig .
[ fig : p1 ] .
indeed , there is a nearly spherical dense core of fully collapsed h - beads ( black colour ) and rather outstretched p - blocks ( gray colour ) hands sticking out of it in a micellar fashion .
now , if we compare the corresponding plots in figs .
[ fig : g1 ] and [ fig : fc]a for @xmath86 vs the arm index , bearing in mind the difference between the theoretical and simulation models and different sizes of considered stars , the agreement between them seems very striking indeed .
we may also observe from fig .
[ fig : g1 ] that the larger is the star ( i.e. greater @xmath1 ) , the larger is its size and the higher is the slope of an increasingly linear increase of @xmath86 with the arm index for the p - beads .
zag pattern for the h - beads is absent in fig .
[ fig : g1 ] because the bead and spring model does not lead to any particular ordering of the globule , contrary to the freely jointed chain model , in which bonds have to be arranged in a special manner consistent with close packing . yet
, the observed behaviour indicates a rather universal character of the subject in hand .
let us now compare partial and total densities in figs .
[ fig : g4 ] and [ fig : fd ] . despite the obvious limitations of the gaussian approximation , in which densities can only be represented by a sum of gaussoids
, we still do find a good general agreement between theory and simulations once we are comparing reduced dimensionless densities @xmath87 vs @xmath48 .
the overall shapes , as well location of peaks and their heights , for all three curves appear to be rather consistent . to avoid any repetition of their discussion already presented in sec .
[ sec : analres ] , we would like to concentrate instead on the distinctions , thereby clearly establishing possible limitations of the theory .
first of all , a sum of gaussoids can never produce a rather flat shape .
thus , the density of the p - beads , @xmath47 , is practically zero in fig . [
fig : g4 ] until almost @xmath88 ( i.e. within the h - core ) , while it can not possibly be so in fig .
[ fig : fd ] being a sum of gaussoids .
apart from that feature , for the h - beads @xmath51 does vanish abruptly , while @xmath47 and @xmath89 do have long tails , for @xmath90 in both figs .
[ fig : g4 ] and [ fig : fd ] . to satisfy the normalisation conditions ,
because of a somewhat different behaviour of @xmath47 near the centre of mass , all densities in the gaussian theory have to be somewhat deformed as compared to their precise shape in the computer simulation .
nevertheless , the density plots seem quite consistent which each other , and particularly so for the total density .
now then , we turn our attention to the outer - h stars . in fig .
[ fig : p2 ] we exhibit a snapshot of the 12-arm star in the stable equilibrium state obtained by a quasistatic change of @xmath18 with attendant long equilibration after every step .
the structure does indeed have a nearly spherical collapsed h - core ( black beads ) , with the p - blocks ( gray beads ) looping out in a daisy like manner .
clearly , the branch point p - bead ( black clubsuit ) has been able to escape outside the core in all of the many analysed members of the ensemble , and it is the centre to which all other p - blocks are attached by harmonic springs .
overall structure is much more compact than in fig .
[ fig : p1 ] and this is also reflected in the graphs in fig . [
fig : g2 ] compared to those of fig .
[ fig : fc]b . in order to avoid over emphasising the good general agreement again , let us instead remark that the branch p - bead is trapped around the centre of mass in the analytical result of fig .
[ fig : fc]b , producing too small values of @xmath86 for neighbouring to it p - beads as opposed to the simulation results in fig .
[ fig : p2 ] . in the latter ,
all p - beads are clearly farther away from the centre of mass than any of the h - beads , and this behaviour seems to make more sense physically .
thanks to the availability of data for different values of @xmath1 , one can also see increasingly better ability of looping of the p - blocks in stars with larger number of arms , so that the @xmath86 plot becomes more and more symmetrically bell shaped on the left .
density profiles in figs .
[ fig : g5 ] and [ fig : fe ] again have consistent overall shapes , peak locations and heights , except that @xmath47 is vanishing for @xmath91 in the simulation results and it can not possibly be so in the gaussian theory for the reason already outlined .
now , let us examine the state obtained as a result of the kinetic process of star collapse after a sudden quench of @xmath92 .
the corresponding snapshot is presented in fig .
[ fig : p3 ] and it does look like a star consisting of @xmath1 p - arms with small h - subglobules at their ends , as predicted by the analytical theory .
this state was found to be a rather long lived metastable one for quenches to @xmath93 the associated nucleation time needed to overcome the barrier for transformation to the equilibrium state of fig .
[ fig : p2 ] was clearly much longer than the monte carlo times practically accessible to our computational resources ( which consisted of a 16-node beowulf pii-400 mhz cluster ) .
however , we have also found that for much deeper quenches the nucleation regime would transform into the spinodal decomposition one , in which the state of fig .
[ fig : p3 ] becomes unstable , so that the kinetics quickly passes through this intermediary to achieve the final equilibrium structure .
looking at the observables such as the mean squared distances and densities as measured from the centre of mass , there is a good agreement of figs .
[ fig : g3 ] and [ fig : fc]b , as well as figs .
[ fig : g6 ] and [ fig : ff ] , even though in the one case a kinetic procedure was used , while a local free energy minimum was used in the other .
we note also that the statistical quality of figs .
[ fig : g3 ] and [ fig : g6 ] is poorer than of similar equilibrium plots presented before for obvious reasons , yet it is quite satisfactory to resolve the shapes well .
the kinetically obtained densities are somewhat broader , which possibly reflects a better ability of p- and h- beads to move and mix within their preferred regions of space .
these observations allow us to firmly conclude that the current metastable state of outer - h stars is indeed a robust metastable state under the considered values of the degree of amphiphilicity @xmath18 , while it becomes an unstable kinetic intermediary for deeper quenches , consistent with the general theory of phase transitions and many of concrete examples of similar behaviour previously seen by us for conformational transitions of other polymeric systems @xcite .
finally , it would be interesting to study the effect of changing the composition ratio of h : p monomers within the arms at a given arms number , for instance @xmath94 .
thus , we have studied the following ratios @xmath95 as well as the homopolymers @xmath96 in addition to the cases of symmetric diblocks @xmath97 studied for both inner - h and outer - h topologies .
first of all , we find that the above general conclusions about the existence of a metastable state for outer - h stars as well as the structures of conformations remain valid in a rather broad range of composition ratios . in fig .
[ fig : as1 ] we present the dependence of the mean squared radius of gyration on the h : p ratio for the three respective structures in poor solvent .
obviously , the size of all three structures decreases monotonically with increasing number of the h - beads .
our conclusion that the outer - h globules are more compact than the inner - h ones remains true here throughout , being most pronounced around small values of @xmath98 , in this case near @xmath99 .
interestingly , the large - size metastable state for outer - h stars exists for all concentrations except very near the two limiting homopolymer cases .
moreover , no other metastable states have been seen in kinetics . in figs .
[ fig : as2],[fig : as3],[fig : as4 ] we exhibit the mean squared distances of monomers from the centre of mass for the inner - h , outer - h and metastable outer - h states respectively .
naturally , the shapes of these curves experience predictable and monotonic shifts , reflecting the changes of the lengths of the h- and p - blocks .
so , @xmath86 for the inner - h structure remains nearly constant within the collapsed h - core , starting to grow characteristically for the p - corona beads with increasing chain index @xmath31 .
this function for the stable state of outer - h stars has a familiar bell
shape for the p - units , forming loops around the collapsed h - core , for which the function remains nearly constant .
and for the metastable state of the outer - h stars we have a steady growth of @xmath86 as long as @xmath31 belongs to the p - block , which changes to a nearly constant behaviour when @xmath31 reaches the h - block
. such types of behaviour are fully consistent with the structures seen in snapshots for the symmetric diblocks .
it is also interesting to follow in more detail the separation of the core monomer @xmath44 from the centre of mass . for the inner - h globules
the core monomer remains very close to the centre of mass similarly to the simple homopolymer star .
the same conclusion is also true for the metastable outer - h structures , whereas for the stable outer - h state @xmath100 is quite large , especially for small @xmath98 ratios due to long loops formed by the p - beads to all of which the core monomer is connected .
in the present paper we have studied the behaviour of block copolymer stars in dilute solution by means of analytical gaussian variational theory and monte carlo simulation in continuous space .
we considered stars with each arm consisting of a diblock chain , one block being formed by hydrophilic ( polar ) monomers or p - beads , and the other one by hydrophobic monomers or h - beads .
moreover , we were interested in symmetric stars either having the h - block connected to the core bead and with the p - block being outside an inner - h star , or having the opposite connectivity an outer - h star . in both cases , we have found that for a sufficiently large degree of amphiphilicity the stable state corresponds to conformations with a dense globule formed by the h - beads , which is surrounded by the p - blocks stretching outwards .
however , because of the intramolecular connectivity , the outer - h stars have a considerably smaller overall size than the inner - h ones . indeed , the p - beads have to return back towards the globule of the h - beads and they are all connected to the same core monomer .
interestingly , for a given degree of amphiphilicity , the size of the h - globule turns out to depend only on the number of h - beads , but it is basically independent of the star topology ( i.e. whether it is an inner - h or an outer - h star ) , and only rather weakly dependent on the number of arms .
the latter dependence is related to the repulsive interactions among the p - beads , whose number increases with the star functionality
. another interesting result of our study was in finding that there is a metastable state for the outer - h stars .
the conformation corresponding to this state looks precisely as a swollen mini star with the h - beads clustering into small subglobules at the ends of the outstretched p - blocks .
the repulsive interactions of the inner p - beads and an entropic barrier prevent the coalescence of these small subglobules into a single globule , as it is the case in the stable state .
importantly , this metastable state was also found by the dynamic monte carlo simulation during kinetics after a sudden quench from the swollen coil to the poor solvent regime .
this kinetic intermediary appears to be a rather long lived metastable state separated by a nucleation stage from the final equilibrium state .
we have also seen that this nucleation regime would transform into a spinodal regime rendering such a state unstable during kinetics after a much deeper quench ( i.e. larger degree of amphiphilicity ) .
one may argue that the existence of such a metastable state would have an obvious implication for the phase diagram of these stars .
indeed , the attractive interactions among the h - beads , which lead to the intramolecular formation of the small droplets at the ends of the arms , would also lead to star aggregation through intermolecular interactions and , eventually , to phase separation .
conversely , in the stable state , the inner globules of two molecules would not aggregate as easily because they are shielded by the outer p - beads exhibiting mutually repulsive interactions , which therefore act as colloidal stabilisers .
we have also studied diblock stars with varying h : p composition ratio and have found that the above general conclusions , such as e.g. the existence of the outer - h metastable state , remain valid for a very broad range of composition ratios .
the family of plots for the mean squared distances of monomers from the centre of mass versus the chain index has been systematically obtained , but it uncovers only very predictable patterns . given a certain technical distinction in the models adopted by the analytical gaussian theory and the monte carlo simulation , as well as expected limitations of the former due to approximations involved , we have found a very good agreement in the results from both techniques .
this is a clear indication that the behaviour we are seeing is indeed universal , and it is not really affected by the analytical approximations or implementation of the simulation .
speaking of small discrepancies between the results , one should mention the trapping of the core p - bead within the h - globule in the stable state of outer - h stars as seen by the analytical theory .
however , all simulation results indicate that this bead can actually escape from the globule , clearly leading to a more energetically favourable situation .
the trapping is possibly enhanced by the small number of degrees of freedom for short lengths of the arms , but in any case is believed to be an artefact of the theory .
perhaps one of the most impressive agreements in the results was obtained for the density profiles , as long as one used the reduced variables @xmath101 vs @xmath48 , with @xmath39 being @xmath41 , @xmath40 , or @xmath42 .
even though the numbers of arms and their lengths differed significantly in both considered cases , the plots in terms of the reduced dimensionless variables showed a robust consistency with each other .
this may seem unexpected given that the density can only be expressed via a superposition of gaussian shaped functions within the analytical theory .
one noticeable discrepancy with the simulation results was found , however , for @xmath47 in the stable globular state of the inner - h and outer - h stars within the range @xmath102 . in that region
, the density of the p - beads goes quickly to zero for @xmath103 in the results of the simulation , showing a perfect @xmath41-beads expulsion from the space occupied by the dense h - subglobule .
this behaviour , however , manifests itself only in a relatively moderate density decrease in case of the gaussian theory , which on the other hand somewhat distorts the overall density shape due to the normalisation requirement .
apart from this understandable difference , the overall shapes , peaks locations , heights and widths are in good agreement between the theory and simulation .
the overall good agreement of the gaussian variational theory with the simulation in application to such complicated a system as a diblock copolymer star , demonstrates the strength of the theory for predicting the details of polymer conformations .
the main idea of the gaussian variational theory is in determining the effective connectivity matrix between all beads which would mimic the true intramolecular interactions in the most optimal way , i.e. minimise the trial free energy , @xmath104 , where @xmath105 and @xmath106 are the trial hamiltonian and free energy associated with it .
the choice of variational variables differs somewhat in the two versions of the theory : it involves products of bond vectors @xmath107 in the milan version , while the set of mean squared distances @xmath108 between monomers in the dublin version .
however , as it is demonstrated in the appendix below , the resulting free energy expressions are in fact the same , up to a trivial additive constant and some rescaling of the parameters of the model . in
so far as the phase diagram and behaviour of simple observables , such as e.g. radius of gyration , mean squared distances and density profiles , are concerned , the gaussian self
consistent approximation turns out to be quite accurate .
it is also clear that the variational theory should become increasingly successful in dealing with polymers with complex intramolecular architecture as compared to linear chain polymers precisely because the conformation of the former polymers is more dictated by the connectivity constraints .
furthermore , the current theory also permits extensions to the realm of kinetics by writing time
dependent equations for instantaneous values of the adopted averaged variables in terms of the instantaneous free energy gradients ( see . e.g. ref .
@xcite ) . such future kinetic studies of copolymers with complex topology present a considerable interest . to illustrate this point , we may mention that folding of even a homopolymer star proceeds considerably faster than of the equivalent linear chain , and this is even more so for an inner - h star vs diblock linear chain . in certain cases of `` random '' copolymers with a specially optimised complex topology
one may achieve a much more efficient and faster folding than for the equivalent linear chain , which would yet produce a globule with an almost identical conformational structure since the connectivity constraints are less relevant within the dense globule .
this may present certain new insights and opportunities for the broad areas of biopolymer folding and perhaps even drug delivery .
speaking of more traditional applications , we believe that the results of this work may facilitate some further experimental research on copolymer stars with a view of their applications such as as viscosity modifiers and coating additives . on the one hand , the existence of distinct stable compact and semi extended conformational states with good colloidal stabilisation properties seems of certain interest .
the presence of the semi
extended metastable state with `` sticky '' properties , moreover , indicates that copolymer stars in solutions at higher concentrations may exhibit a spectacular phase behaviour .
while this is interesting , the studies of aggregation phenomena within the present very detailed treatment are not particularly justified due to considerable computational expenses . a somewhat more coarse grained approach , both analytically and in terms of the monte carlo model , may lead to a faster progress in this direction .
we expect that appropriate techniques for such analysis could soon be developed and that the corresponding complex phenomena will be thoroughly studied within the next few years
. finally , we would like to express a hope that the theoretical and computational study of macromolecular solutions involving polymers with increasingly complex design , as well as the collaboration between our two groups , will continue in the future .
the authors are thankful to professor giuseppe allegra , professor angelo perico and dr guido raos for interesting discussions .
one of us ( e.t . ) would like to thank gavin mccullagh and ronan connolly for their work on the related gui environment and gaining preliminary insights into the problem during the course of their undergraduate projects .
this work was supported by grant ic/1999/01 from enterprise ireland , which has helped to develop a fruitful collaboration between the dublin and milan groups .
comparison of the results from the theory and simulation using different models of connectivity in this paper show that as far as the universal properties of polymers are concerned one could use either of the two models : freely jointed or bead and spring one .
since the connectivity term ( [ cmc : hamil ] ) is the only one which is affected by the molecular topology , it would suffice to discuss the theory for a linear polymer chain , in which one has to change the connectivity matrix appropriately .
the purpose of this appendix is to give some mathematical details on the analytical equations employed , but as an added bonus to demonstrate the equivalence of the equilibrium free energy expressions used independently by the two groups .
thus , let us establish an explicit relationship between the free energy of the gsc method in ref .
@xcite and that of refs .
@xcite .
first of all , monomer coordinates can be written for a linear chain via the bond vectors as follows , @xmath109 , if we choose the reference point as , @xmath110
. then , the mean squared distances between monomers @xmath31 and @xmath53 are written as @xmath111 therefore , the entropic term ( eq . ( 7 ) in ref .
@xcite ) in units of @xmath112 , @xmath113 where we have discarded a constant independent of the variational variables and the matrix @xmath114 defined as , @xmath115 can be rewritten as follows .
its first part becomes the harmonic springs `` energy '' ( eq .
( 10 ) in ref .
@xcite ) , @xmath116 which in the latter form would also be valid for stars and any other topology as it is precisely the first term in eq . (
[ cmc : hamil ] ) employed in sec .
[ subsec : cmc ] .
now , let us consider the determinant of the matrix @xmath114 , where we have scaled out the constant @xmath117 , @xmath118 obviously , the determinant of a matrix does not change if one adds a column or a row to any other columns or rows .
thus , by adding the first row to the second , then the new second to third and so on , we obtain , @xmath119 by repeating the same procedure for the columns , likewise we conclude that @xmath120 which in turn agrees with eqs .
( b3 ) and ( b5 ) in ref .
@xcite upon fixing the reference point at @xmath110 . finally , the virial expansion of the excluded volume energy ( eq .
( 10 ) in ref .
@xcite ) , @xmath121 expands into the following explicit formulae for the determinants appearing in the two and three body interaction terms , @xmath122 these are precisely the expressions for the two body and three body interaction energy terms as in eqs .
( 50,51 ) of ref .
the positiveness of the first determinant requires all mean squared distances to be positive , @xmath123 , which is true as this is the average of a quantity squared .
the positiveness of the three body determinant in eq .
( [ gsc : det3b ] ) , on the other hand , is insured by the triangle inequality , @xmath124 it is interesting to note that the quantity @xmath125 characterises the volume of the phase space occupied by @xmath126 particles if the mean squared distances between all pairs are specified .
moreover , the logarithm of this characteristic volume of all @xmath29 particles is equal to the variational entropy of the system up to a trivial constant , @xmath127 this completes establishing the equivalence of the variational free energy expressions between the methods one which is based on bond
vectors @xcite and another which is based on the monomer coordinates and mean squared distances between monomers @xcite . | @xmath0 summary : we study conformations assumed by single diblock star copolymers in a poor solvent by means of the gaussian variational theory and monte carlo simulation in continuous space .
cases of stars with internal and external hydrophobic blocks are analysed .
while in the former case the collapsed state has an obvious micellar shape , the latter case exhibits two nontrivial conformational structures .
apart from the equilibrium state of a globular hydrophobic core with hydrophilic daisy loops , one also finds here a metastable state of outstretched hydrophilic blocks with hydrophobic subglobules at their ends .
such a state appears to be rather long lived during the kinetics of collapse of a swollen star .
the plots of monomer densities and other observables computed by both techniques are found to be in good agreement with each other . |
( ct ) is a phenomenon of perturbative qcd ( pqcd ) , whose characteristic feature is that small color - singlet objects interact with hadrons with small cross sections @xcite .
cross section for the interaction of such small object , or small size configuration ( ssc ) , with a hadron target has been calculated in qcd using the factorization theorem @xcite .
these qcd calculations confirmed the hypothesis of f. low @xcite on smallness of the cross section for the interaction of ssc with a hadron , if the gluon density in the hadron is not very high ( moderately small @xmath5 ) .
they also predict a related phenomenon of _ color opacity _ when the gluon density becomes very large ( at small @xmath5 ) and ssc interacts with the hadron with large cross section @xcite .
the prerequisite for observing ct is to select a sample containing ssc s via a hard process ( i.e. with large @xmath6 , high @xmath7 , or large produced mass ) . to suppress non - perturbative ( not ssc ) background different additional restrictions
should be imposed depending on the process .
for instance for hard exclusive @xmath8 leptoproduction , in addition to large @xmath0 , selection of the longitudinally polarized mesons is required .
to measure the ssc - nucleon cross sections one should study absorption of the ssc propagating through nuclear matter . in order to clearly observe ct
it is necessary that the ssc lives long enough to traverse distances larger than the size of the target nucleus .
another requirement is that ssc stays small while propagating through the nucleus .
these requirements , which are quantified in terms of the coherence length and the formation length , are discussed later .
various processes have been proposed to study ct phenomenon . * coherent vector meson ( @xmath9 , @xmath10 , @xmath11 ) production at small @xmath12 . for near - forward
( @xmath13 ) coherent production and complete ct , one expects that the cross section for the nuclear target is related to that for the nucleon target by @xmath14 gluon shadowing / antishadowing in nuclei is neglected in this formula , but for moderately small @xmath15 this effect is expected to be small . usually in experiments the @xmath12-integrated coherent cross sections are measured , for which ct predicts an approximate @xmath16 dependence . using more realistic nuclear form factors one predicts @xmath17 @xcite .
+ otherwise , if the color - singlet objects , while propagating in nuclear matter , interact with nucleons with large cross sections , the expected @xmath18-dependence is weaker ; @xmath19 for the cross sections comparable or larger than the pion - nucleon cross section . * incoherent ( quasi - elastic ) vector meson ( @xmath9 , @xmath10 , @xmath11 ) production on nuclei . for complete ct and neglecting effects of the gluon shadowing / antishadowing in the nuclei one expects @xmath20 here , for the nuclear target the meson @xmath21
is produced on a single nucleon of the nucleus and @xmath22 denotes the system of spectator nucleons .
* coherent or incoherent production of excited vector meson states @xmath23 or @xmath24 .
ct predicts the same @xmath18 dependence of @xmath9 and @xmath23 , or @xmath10 and @xmath24 .
this contrasts with the naive expectations whereby one may expect larger absorption for excited mesons since they are larger .
* coherent diffractive dissociation of hadrons or photons into high @xmath25 di - jets .
such process probes the small transverse - size component of the projectile wave - function as well as ct effects . for ct
the @xmath12-integrated cross section of coherent diffractive production of high @xmath25 di - jets has the same @xmath18-dependence as for the processes ( a ) . using more realistic wave functions
one predicts for asymptotically high energies @xmath26 dependence , where @xmath27 is in the range 1.45 1.60 , depending on @xmath25 @xcite .
* coherent vector meson production on light nuclei ( deuteron , helium ) in the large @xmath12 range , where the effects of double ( multiple ) scattering are important .
ct will suppress the double ( multiple ) scattering contribution to the differential cross section @xmath28 @xcite .
* @xmath18-dependence of the fraction of large rapidity gap ( diffractive ) events .
ct predicts an @xmath18 independent fraction ; otherwise it will grow with @xmath18 @xcite .
* large @xmath0 quasielastic @xmath29 scattering on nuclei , @xmath30 . in the ct
limit the cross section will be proportional to @xmath18 , as for process ( b ) @xcite . *
large @xmath12 quasielastic @xmath31 scattering on nuclei , @xmath32 .
also for this process , for complete ct the expected @xmath18 dependence is like for process ( b ) @xcite . in searches for ct
a commonly used quantity , measured in experiments , is the _ nuclear transparency _
@xmath33 which is the ratio of the cross section per nucleon for a selected process on a nucleus @xmath18 to the corresponding cross section on a free nucleon . for the incoherent processes
ct predicts @xmath34 independently of @xmath18 . for the large absorption in the nuclear matter
@xmath35 will be smaller than unity and will decrease with @xmath18 .
although the nuclear transparency could be defined also for coherent processes , usually cross sections for coherent production on different nuclei are compared directly . as mentioned above , for @xmath12-integrated coherent cross sections ct predicts @xmath36 , whereas for a larger absorption the @xmath18-dependence is weaker . in the following discussion we concentrate on exclusive vector meson production ( vmp ) , the processes ( a)(c ) , which could be studied at the compass experiment @xcite .
exclusive meson production on the nucleon , free or bound in the nucleus , can be viewed as proceeding according to the diagram shown in fig .
kinematic variables used in this paper are listed in table [ kinvar ] .
.kinematic variables used in the text .
[ kinvar ] [ cols= " < , < " , ]
in this section we compare compass capabilities to demonstrate ct to those of previous experiments in which exclusive @xmath8 leptoproduction on nuclear targets was studied .
we concentrate on this reaction , as among different exclusive vmd channels it has the largest cross section . for the comparison we have selected experiments which covered @xmath0 range extending to large values , bigger than 2 gev@xmath1 .
this condition is satisfied for the following experiments : hermes @xcite , nmc @xcite and e665 @xcite . the published results from the hermes experiment concern the incoherent exclusive @xmath8 production on @xmath37 , @xmath38 , @xmath39 and @xmath40 targets .
the electron beam energy was 27.5 gev and the covered @xmath0 and @xmath41 ranges are @xmath42 and @xmath43 fm .
like in all fixed - target experiments with lepton beams , the covered kinematic range is such that for a given @xmath0 the average value of @xmath41 decreases with increasing @xmath0 .
the highest values of the hermes data where one may expect to observe the onset of ct are correlated with the small values of @xmath41 below 2 fm .
the increase of the nuclear transparency observed for @xmath40 is explained as due to the reduced coherence length , without resorting to ct .
wider kinematic ranges were coverd by experiments using high - energy muon beams .
nmc has published the data on coherent and incoherent exclusive @xmath8 production on @xmath38 , c and ca targets .
the muon beam energy was 200 gev and the data cover the ranges @xmath44 and @xmath45 fm .
for the nmc kinematic range , the large values of @xmath41 which exceed the diameter of the largest target nucleus ( ca ) , @xmath46 fm , correspond to @xmath47 . in principle , in this range it is possible to study @xmath0 dependence of nuclear absorption in order to observe ct not obscured by effects of small @xmath41 .
however , due to the moderate statistics of the data , such detailed analysis was not possible for nmc .
the most favourable kinematical conditions were realised in the e665 experiment due to the high muon beam energy of 470 gev .
the experiment has published the data on incoherent exclusive @xmath8 production on @xmath37 , @xmath38 , c , ca and pb targets .
the data were taken in a wide @xmath0 range , including also very small values @xmath48 , and in a wide range of @xmath41 , @xmath49 fm . at small and moderate @xmath0 values , @xmath50 gev@xmath1 , which correlate with @xmath41 values exceeding the diameter of the lead nucleus , precise measurements of the nuclear transparency were obtained .
they indicate strong nuclear absorption . at the highest @xmath0 bin , @xmath51 , the nuclear transparency increases in qualitative agreement with ct .
however , the data in this bin correspond to a wide range of @xmath41 , @xmath52 fm , and therefore could be affected by effects due to the reduced coherence length .
the low statistics in the large @xmath0 bin did not allow more detailed studies .
the compass data at medium and large @xmath0 ( ft trigger ) will cover the kinematic range similar to that of the nmc data .
the expected statistics for the carbon target will be about 2 orders of magnitude higher than that of the nmc data for the same target .
this increase is in particular due to higher beam intensity and larger acceptance in the compass experiment . similarly like in the e665 experiment compass
will use also lead target and will extend measurements to small @xmath0 .
the statistics of compass data will be significantly higher also than that of e665 .
for example , for incoherent events at @xmath51 the ratio of nuclear transparencies for lead and carbon will be determined with accuracy better than 3% , comparing to about 30% for e665 . another more sensitive method to study nuclear effects in exclusive @xmath8 production , the one by using coherent production ,
will be employed in compass as well . due to large statistics
splitting of compass data in several @xmath0 and @xmath41 bins as well as the selection of events with longitudinal or transverese @xmath8 polarization will be possible .
in addition to the @xmath8 production , we will study exclusive production of @xmath9 , @xmath11 , @xmath53 and @xmath54 and their ratios in order to demonstrate ct . for these channels
the analysis will be similar to that for @xmath55 , but the rates will be lower due to the smaller production cross sections .
the estimates of the required period of data taking and of the target thicknesses are subject of further analysis .
we consider also a possibility to study ct via coherent production of hadron pairs with large relative transverse momenta .
the experimental procedure would be similar to that used in the fermilab experiment e791 @xcite , in which ct was observed in coherent diffraction of pions into high-@xmath25 di - jets .
we will select a sample of events with high-@xmath25 hadron pair and an additional requirement that all observed charged hadrons carry at least 90% of the energy of @xmath56 .
an estimate of @xmath12 will be obtained using momenta of all measured charged hadron tracks .
the expected resolution in @xmath12 , which will be crucial for selection of the coherent sample , will be determined from a dedicated simulation and reconstruction of events with high-@xmath25 hadron pairs .
finally we mention few experimental aspects which have to be further investigated .
+ here we propose to position the nuclear targets downstream of the polarized target .
then the ct physics could complement the standard compass program by using the downstream nuclear targets simultaneously with the polarized target .
similar target configuration was already used during the 1985 running of the emc experiment , when the polarized target was used together with several nuclear targets situated downstream of pt .
the effect of the downstream nuclear targets on the hadrons coming from the polarized target should be small .
for instance , for the reconstructible tracks of hadrons from pt , only about 4% of slow pions from @xmath57 decays and about 1% of hadrons from @xmath58 decays will be in the acceptance of the proposed nuclear target @xcite .
for the scattered muon coming from pt and passing through the nuclear target , the additional multiple coulomb scattering should result in a small increase of the errors of the reconstructed position of the event vertex .
assuming a dedicated run with unpolarized targets , different target configurations for ct studies are possible .
with more available space one could envisage targets at least about 4 times thicker than the thin targets presented earlier .
thus the four - fold reduction of the data taking time will be possible . with an even larger amount of available beam time
, one could expose more nuclear targets of different materials , and in addition the liquid hydrogen or deuterium targets .
the latter ones are also important for measurements of _ off - forward parton distributions _
( also referred to as _ skewed parton distributions _ or _
generalized parton distributions _ ) @xcite . for a dedicated run with unpolarized targets
, one could take advantage of a possibility of adding a recoil detector .
this is crucial for certain processes , e.g. deep virtual compton scattering ( dvcs ) , but in any case it will help to reduce the non - exclusive background for any exclusive process .
we have simulated and analysed exclusive @xmath59 muoproduction @xmath60 and @xmath61 at the compass experiment using thin nuclear targets of carbon and lead . for muon beam energy of 190 gev and a trigger for medium and large @xmath62 ,
the covered kinematic range is @xmath63 and @xmath64 .
good resolutions in @xmath65 , @xmath66 , @xmath12 ( @xmath67 ) and @xmath68 are feasible . an efficient selection of coherent or incoherent events is possible by applying cuts on @xmath67 . in order to obtain the samples of events initiated with a probability of about 80% by either @xmath69 or @xmath70 ,
the cuts on the @xmath59 decay angular distribution of @xmath71 will be used .
the search for ct could be facilitated by using the events with @xmath72 values exceeding the sizes of the target nuclei .
the fraction of such events is substantial and the covered @xmath62 range seems sufficient to observe ct .
we showed high sensitivity of the measured ratio @xmath73 of nuclear transparencies for lead and carbon for different models of nuclear absorption .
good statistical accuracy of the measured @xmath73 may be achieved already for one year of data taking .
these measurements , taken at different @xmath62 intervals , may allow to discriminate between different mechanisms of the interaction of the hadronic components of the virtual photon with the nucleus . in conclusion
, the planned comprehensive studies of exclusive vector meson production on different nuclear targets at the compass experiment should unambiguously demonstrate ct .
the authors gratefully acknowledge useful discussions with l. frankfurt . we thank also s. kananov for discussions on the zeus exclusive @xmath8 data .
the work was supported in part by the israel science foundation ( isf ) founded by the israel academy of sciences and humanities , jerusalem , israel and by the polish state committee for scientific research ( kbn spub nr 621/e-78/spub - m / cern / p-03/dz 298/2000 and kbn grant nr 2 p03b 113 19 ) .
one of us ( a.s . ) acknowledges support from the raymond and beverly sackler visiting chair in exact sciences during his stay at tel aviv university .
et al _ , _ compass , a proposal for a common muon and proton apparatus for structure and spectroscopy _ ,
+ cern / spslc/96 - 14 , spsc / p297 ( 1996 ) and cern / spslc 96 - 30 ( 1996 ) , + http://wwwcompass.cern.ch/compass/proposal/welcome.html .
frankfurt , m.v .
polyakov and m.i .
strikman , _
n@xmath74 dvcs , exclusive dis processes and skewed quark distributions in large n@xmath75 limit _ ,
workshop on jefferson lab .
physics and instrumentation with 6 - 12 gev beams and beyond , newport news , virginia , june 1998 , hep - ph/9808449 .
a. tripet , presented at the 7th international workshop on deep inelastic scattering and qcd , zeuthen , germany , 19 - 23 april 1999 ; proceedings , j. blmlein and t. riemann , nucl .
b , proc . suppl .
* 79 * ( 1999 ) 529 .
x. ji , phys .
* 78 * ( 1997 ) 610 ; phys . rev . *
d 55 * ( 1997 ) 7114 ; + a.v .
radyushkin , phys . lett .
* b380 * ( 1996 ) 417 ; phys . rev . *
d 56 * ( 1997 ) 5524 ; + j.c .
collins , l. frankfurt and m. strikman , phys .
d 56 * ( 1997 ) 2982 . | we examine the potential of the compass experiment at cern to study color transparency via exclusive vector meson production in hard muon - nucleus scattering . it is demonstrated that compass has high sensitivity to test this important prediction of perturbative qcd .
2(c)^2 ps .
|_| 2r^2 2@xmath0 2c2gev@xmath1/@xmath2 2 ^ 2 # 1#2#3#4#1 * # 2 * , # 3 ( # 4 ) 2 2pigamma taup-2671 - 2001 + 12 june 2001 + feasibility study + @xmath3 sotan institute for nuclear studies , ul . hoa 69 , pl 00 - 681 warsaw , poland , @xmath4 school of physics and astronomy , r. and b. sackler faculty of exact sciences , + tel aviv university , 69978 ramat aviv , israel _ e - mails : [email protected] , [email protected] , [email protected] , [email protected]_ |
the birth of functional renormalization group methods is tightly connected to the investigation of scaling properties of the effective potential in statistical field theories with @xmath0 or @xmath3 symmetries @xcite , and it represents its first non - trivial application @xcite . over the years
the interest in properties that can be extracted from the renormalization group flow of the potential in a local approximation has been revitalized at various stages @xcite .
the effective potential of a @xmath0-symmetric scalar field theory has been studied with various techniques to unveil the underlying universality classes , which include the ising model .
the critical exponents have been determined to high accuracy within polynomial expansions @xcite and non - polynomial solutions have been studied numerically by applying scaling solutions methods @xcite . recently
, even global solutions have been constructed with very high numerical precision @xcite . in the past few years
, the scaling solutions approach was adopted to analytically continue various parameters which enter the renormalization group flow , including the dimensionality of space @xcite , the number @xmath4 of fields @xcite or both @xcite .
the analytic continuation of these parameters offers new insights into properties of the underlying statistical models , such as their emergence on spaces of fractional dimensions , the occurrence of the mermin - wagner theorem and constraints on the existence of the various multicritical models . in this paper
we extend the scaling solution approach in three prominent ways .
first , we will carefully lay down and discuss the theoretical mechanism with which a scaling solution occurs for a specific statistical model , which has often been overlooked in previous works .
this occurrence will be illustrated mostly through the example of the three dimensional ising model .
second , we will introduce two methods to compute the critical exponents of a model : a shooting approach ( which has been used previously in @xcite ) to calculate critical exponents in three and two dimensions ) and the novel pseudo - spectral method with a slac derivative @xcite .
these two methods treat the boundary conditions for the deformations of a critical potential in different ways ; thus , they are complementary for the approach as will be shown by applying them to the known ising example .
finally , we systematically apply the scaling solution approach to the supersymmetric @xmath5 wess - zumino model in both two @xcite and three dimensions @xcite , for which we compute various critical exponents for its universality classes .
we also study the supersymmetric model on fractional dimensions introducing a novel analytic continuation .
the paper is organized as follows : the remaining part of section [ section_intro ] introduces the local potential approximation and the functional renormalization approach , sect .
[ section_simple_scalar ] introduces the scaling solution approach and applies it to various universality classes in two and three dimensions , sect .
[ section_wz_model ] introduces the wess - zumino model and shows the application of the scaling solution method to the supersymmetric analogues of the universality classes of the previous section , sect .
[ section_critical_dimensionalities ] shows how the universality classes emerge at certain critical dimensions when varying the dimensionality of the system , and finally sect .
[ section_discussion ] briefly discusses our results .
the appendixes are dedicated to further clarifying some properties of the renormalization group flow and the methods used in this work . for clarity purposes we introduce all the basic concepts used in the present paper adopting the illustrative example of a simple @xmath0-invariant scalar field on @xmath6 . this introduction can and will be straightforwardly generalized to the inclusion of supersymmetry later in the corresponding sections .
readers that are experts in the techniques of wilsonian and functional renormalization are welcome to skip the remainder of the present section .
the local potential approximation ( lpa ) is a non - perturbative truncation of the effective action of a system which includes all the relevant operators in a low energy derivative expansion @xcite . within this approximation ,
higher derivative operators are neglected but all local interactions are included in a non - perturbative fashion .
the lpa of a simple @xmath0-invariant scalar @xmath7 in @xmath8 dimensions thus truncates the effective action of the system to @xmath9 = \int { \rm d}^d x \left\ { \frac{z}{2}\partial_\mu \varphi \partial^\mu\varphi + v(\varphi ) \right\}\,,\ ] ] where we included a wavefunction renormalization constant @xmath10 for the field @xmath7 and the local potential @xmath11 .
more precisely , the truncation represents an extension of the original lpa of a simple scalar that goes under the name of improved lpa ( a.k.a .
lpa ) , as it includes a non - trivial wavefunction renormalization for @xmath7 and thus allows the field to have a non - canonical scaling in the guise of an anomalous dimension for the renormalized field @xmath12 .
we study the renormalization group flow of the scalar models using functional renormalization group ( frg ) methods . in this approach
the path - integral of the system is constructed as @xmath13 } = \int { \cal d}\chi \ , { \rm e}^{-s[\chi]-\delta s_k[\chi ] + ( \chi-\varphi)\frac{\delta\gamma_k}{\delta\varphi}}\,,\ ] ] where @xmath14 $ ] is a bare action for the microscopic field @xmath15 and @xmath16 $ ] is a cutoff that suppresses the propagation of the infrared modes relative to a given reference scale @xmath17
. the infrared cutoff should be quadratic in the field @xmath18 = \frac{1}{2 } \int { \rm d}^d x \ , { \rm d}^d x ' \ , \chi(x ) { \cal r}_k(x - x ' ) \chi(x')\,,\ ] ] and its kernel @xmath19 suppresses the propagation of the infrared modes by directly modifying the bare propagator of the theory .
the dependence of on the kernel @xmath19 encodes the scheme - dependence of the method ; thus , different kernels implement different coarse - grainings .
however , the kernel has to satisfy in the limit @xmath20 the property @xmath21 , which ensures that reduces to the standard path - integral of the ( full ) effective action @xmath22 $ ] of the system when the reference scale @xmath17 is taken to zero .
the functional @xmath23 $ ] appearing on the rhs of is known as the effective average action ( eaa ) of the model . by construction
it reduces to the full effective action @xmath24=\gamma[\varphi]$ ] in the infrared limit @xmath20 .
furthermore , for sufficiently regular cutoff kernels it can be shown that an ultraviolet limit can be constructed @xmath25=s_\lambda[\varphi]$ ] , in which @xmath26 is an ultraviolet cutoff and @xmath27 $ ] is the ( possibly divergent ) bare action of the model at the uv - scale @xmath26 . while the path integral is generally not free of ultraviolet divergences and , in principle , requires renormalization , it is possible to show that the flow of @xmath23 $ ] with respect to the scale @xmath17 is finite .
therefore , rather than achieving the renormalization of the theory through the inclusion of counterterms , it is customary to use to derive the exact and finite frg equation @xmath28 = \frac{1}{2 } { \rm tr } \left(\gamma^{(2)}_k[\varphi]+{\cal r}_k\right)^{-1}k\partial_k { \cal r}_k\,,\ ] ] and use it to investigate the renormalization of the model .
the flow encoded by determines a velocity field in the space of all functionals of the fields entering a model under investigation .
the lpa can be applied easily by projecting both sides of to a @xmath17-dependent generalization of which reads @xmath29 = \int { \rm d}^d x \left\ { \frac{z_k}{2}\partial_\mu \varphi \partial^\mu\varphi + v_k(\varphi ) \right\}\,.\ ] ] the anomalous dimension of the field is @xmath30 . in order to read off the renormalization group flow of the couplings in we
choose the prescriptions @xmath31
\right|_{\varphi={\rm const.}}\ , , \\ & \eta = -\frac{1}{z_k } k\partial_k \frac{\partial}{\partial p^2 } \left.\frac{\delta^2 \gamma_k[\varphi]}{\delta \varphi_p \delta\varphi_{-p}}\right|_{\varphi=\varphi_0,\,p^2=0 } \end{split}\ ] ] which must be applied to the right hand side of .
various remarks are in order to explain the meaning of : the flow of the local potential is defined as the ( normalized ) flow of the eaa in the limit of constant field content , as is usual in quantum field theoretical applications .
the field @xmath32 represents the fourier transform of @xmath33 with respect to the momentum @xmath34 ; thus , the anomalous dimension @xmath35 is extracted quite naturally from the coefficient of the @xmath36 term in the two - point function belonging to @xmath23 $ ] .
since the truncation includes only a field - independent wavefunction renormalization constant , it is necessary for consistency to project the two - point function of @xmath23 $ ] to a distinguished field configuration @xmath37 . in applications @xmath37
is chosen in a physically meaningful way as the minimum of the effective potential defined by @xmath38 .
the configuration @xmath37 thus plays the role of the order parameter of the @xmath0-symmetry @xmath39 , and the possibility that it has non - zero values represents a clear separation from standard perturbation theory .
let us now define the modified momentum - space propagator at constant field @xmath40 where @xmath41 is @xmath42 times the fourier transform of the cutoff kernel in coordinate space . with the modified propagator we can give a compact evaluation of based on a general cutoff @xmath43 here
, the prime denotes the derivative with respect to the argument @xmath44 , whereas elsewhere it is the derivative with respect to the ( constant ) field .
we can give explicit expressions for the flow equation for the effective potential and the anomalous dimension after specifying the cutoff function . while , in principle , any choice compatible with the properties of a suitable cutoff is possible , for the purpose of this paper it is useful to choose one that allows for an explicit calculation of the flow .
we choose the popular optimized cutoff and explicitly factor out the wavefunction renormalization of the field , so that in momentum space the kernel of reads @xmath45 the optimized cutoff is a distribution and therefore the momentum space integrals must be treated with care . in requires a normalization for the integral of the product of the heaviside step function @xmath46 and its first derivative @xmath47 .
given a regular test function @xmath48 we normalize @xmath49 this normalization is consistent with the limit of the same integral evaluated for any sequence of smooth functions approaching @xmath46 . ]
let us also introduce the dimensionless renormalized field @xmath50 , which is the natural argument of the dimensionless renormalized potential @xmath51 together @xmath52 and @xmath53 are the appropriate variables to investigate the scaling properties of the system , and in fact we shall use them for the remainder of this section . for notational simplicity
we also drop both the labels and the overline notation , so the explicit evaluation of the flow in the new variables takes the form @xmath54+{\cal f}_{v,\eta}[v'']\ , , \\ & { \cal s}_{v,\eta}[v , v';\varphi]\equiv - d v(\varphi)+\frac{d-2+\eta}{2}\varphi v'(\varphi)\ , , \\ & { \cal f}_{v,\eta}[v'']\equiv c_d \frac{1-\frac{\eta}{d+2}}{1+v''(\varphi)}\ , , \end{split}\ ] ] in which we defined the dimension - dependent geometric constant @xmath55 . we split the flow of the potential in two parts with different origins .
the functional @xmath56 $ ] represents the scaling part of the flow and it originates from the change of dimensionful variables to dimensionless ones .
its form is independent from the cutoff choice and therefore from the scheme of computation .
the functional @xmath57 $ ] , instead , encodes the nontrivial features of the flow and depends on the cutoff .
we now compute from the anomalous dimension explicitly to obtain @xmath58 where again we assume @xmath37 to be the expectation value of the order parameter of the system . for later convenience
we also give the result of an alternative computation of the anomalous dimension , which comes as the limit @xmath59 of the anomalous dimension of the goldstone modes of the @xmath3-model @xmath60 this definition has often and successfully been used in place of @xmath35 on the basis of physical arguments .
thus , we include it in our analysis and refer to it as goldstone anomalous dimension .
the reason why @xmath35 and @xmath61 are different is not because the limit @xmath62 of the @xmath3-model is discontinuous , but rather because @xmath35 represents the limiting value of the anomalous dimension of the massive higgs field in the broken @xmath3 phase .
let us now discuss some general property of the flow of the potential .
the computation can be manifestly analytically continued to any non - integer dimensionality , and for the present work we are mostly concerned with the interval @xmath63 which is bounded by the lower and upper critical dimensions of the @xmath64 model .
note that the flow of the potential depends _ parametrically _ on the anomalous dimension given in or .
nevertheless , for @xmath65 we can extract some general properties of the flow that are insensitive to the values of @xmath35 .
the flow is a pde representing the evolution of the potential with respect to the scale @xmath17 .
the @xmath17-independent ( stationary ) solutions are the critical solutions ( fixed points ) of the flow , and they correspond to critical models of the corresponding field theory .
this means that critical solutions solve the non - linear ode @xmath66+{\cal f}_{v,\eta}[v'']=0\,,\ ] ] which is called a fixed point equation .
the function @xmath67 can be explicitly inverted to cast the ode in a standard form @xmath68\,]\,.\ ] ] while the explicit form of this ode depends on the cutoff , it is easy to prove that for any cutoff @xmath69 \sim 1/x\,,\,\,\,\ , { \rm for}\,\,\,\ , x \to \infty\,,\ ] ] which implies @xmath70\,,\,\,\,\ , { \rm when}\,\,\,\ , { \cal s}_{v,\eta } \sim 0\,.\ ] ] in other words the second derivative of the potential is expected to diverge at points @xmath71 at which the scaling part becomes zero .
these points can be determined implicitly by the equation @xmath72 as @xmath73 it is possible to show that any attempt to integrate the fixed point equation starting from initial conditions at @xmath74 will stop ( with probability one ) at a singularity located at @xmath71 at which @xmath75 diverges , but @xmath76 and @xmath77 stay finite @xcite .
in addition , if @xmath78 then @xmath76 and @xmath77 have the same sign ; similarly , if @xmath79 they have opposite signs . ideally , we are interested in global analytic solutions of , which thus avoid the appearance of singularities such as for finite values of the field @xmath71 , and we hope that these global solutions can be obtained by numerically integrating the ode with fine tuned boundary conditions . in practice , it turns out that in the process of integrating singularities are the rule and not the exception . to investigate the singular solutions we first choose a suitable parametrization . since is a second order ode ,
any solution can be characterized by the two initial conditions @xmath80 and @xmath81 . one of the conditions is , however , fixed by the requirement of @xmath0-invariance to be @xmath82 , thus ensuring the even parity of the solutions .
it proves convenient to parametrize the second condition using the value @xmath83 rather than @xmath80 .
we thus choose , as an initial condition @xmath84 , which implies that @xmath85 which makes explicit the fact that @xmath86 is a singular point for the ode .
this comes from the more general property that , for any cutoff , the flow is well defined whenever @xmath87 . with the exception of the trivial case @xmath88 , it is possible to observe numerically that almost all solutions that can be considered within @xmath89 diverge at a singular point @xmath71 defined by whose position depends on @xmath90 .
we therefore denote @xmath91 the location of the singularity as a function of @xmath90 .
the second derivative @xmath92 diverges at field values for which the scaling part @xmath93 vanishes . at any maximum or minimum of the potential with positive @xmath94
the scaling part is negative and at any extremum with negative @xmath94 the scaling part is positive . since for @xmath95
the scaling part is negative at @xmath74 and since the potential ( of a stable theory ) is positive for large values of the field we conclude that _ every globally defined fixed point solution @xmath96 is positive_. if @xmath94 takes negative values for certain fields , then it would have a minimum @xmath97 with negative @xmath94 and the scaling part would change sign between @xmath74 and @xmath97 .
this in turn implies that the scaling part would vanish at some field @xmath98 or that the potential would be singular at @xmath91 .
critical values of the parameter @xmath90 can be detected as discontinuities of either @xmath91 or its first derivative @xmath99 . in general , these are directly recognizable when plotting the function @xmath91 as either jumps or spikes . in the universality classes of the scalar model they always appear as jumps of @xmath91 .
let us illustrate how the critical solutions are detected in practice by solving in the case of @xmath100 .
we begin by numerically integrating with initial condition as a function of the parameter @xmath90 , with both formulas specialized to the case @xmath100 and @xmath101 . for almost all values of @xmath90 ,
the numerical integration comes to an end at a singularity @xmath91 defined by and whose value depends on @xmath90 as shown in fig .
[ spike3dplot ] . as a function of @xmath90 by solving for @xmath100 and @xmath101 .
, width=302 ] the plot of @xmath91 reveals various salient features .
the singularity located at the point @xmath86 is due to our choice for the parametrization of the initial condition and simply represents an infinite solution @xmath102 .
the singularity located in the center at @xmath103 is the gaussian solution which trivially extends to all values of the field because the potential is identically constant @xmath104 . on the right of the plot
the function @xmath91 slowly increases to infinity where another trivial solution , representing an infinite mass limit is located . the interesting feature of fig .
[ spike3dplot ] is the singularity located on the left of the gaussian one at @xmath105 , which is characterized by a marked singular behavior of both @xmath91 ( a jump ) and its first derivative with respect to @xmath90 ( a spike ) .
we can try to interpret the meaning of this singularity by observing the behavior of the solution in the vicinity of the critical value @xmath105 as shown in fig .
[ discontinuity ] . and @xmath106 .
, width=302 ] the solutions with @xmath90 slightly on the left of @xmath107 display a minimum and end at the singularity with positive values of @xmath108 and @xmath109 .
those with @xmath90 slightly on the right of @xmath107 end with negative values of @xmath110 and @xmath111 .
the simultaneous change of signs in is responsible for the sudden increase of @xmath91 while crossing @xmath112 from left to right as seen in fig .
[ spike3dplot ] . the left side of the non - trivial singularity in fig .
[ spike3dplot ] displays solutions possessing a single non - trivial minimum before terminating in the singularity
. therefore these are solutions with a non - trivial value of the order parameter @xmath37 such that @xmath113 . on the right of the singularity the solutions sharply bend to negative values and , when following @xmath90 to zero , they continuously connect to the gaussian solution .
it is thus tempting to define the critical solution as the one obtained from the limit @xmath114 _ from the left_. the value @xmath107 has to be determined numerically and , within a ten digits precision , takes the value @xmath115 a closer look at the numerically determined solution reveals the critical properties expected for the potential of the ising - class in three dimension as shown in fig . [ criticalsol ] . .
, width=302 ] the critical potential in this figure has been calculated with the same precision by integrating the ode using @xmath101 as an input .
we refer to the anomalous dimension appearing parametrically in as @xmath116
. the consistency of the lpa can be further tested by using the solution itself to compute the anomalous dimension through the corresponding formula .
the new result for @xmath35 will be an output of the potential s determination and in general will be different from the input parameter .
we thus call the new anomalous dimension @xmath117 . in the case of the potential of fig .
[ criticalsol ] the minimum is located at @xmath118 and the use of yields @xmath119 , thus outlining that , strictly speaking , in the lpa the anomalous dimension is not determined .
the consistent determination of @xmath35 required by the framework of the lpa can be achieved by constructing an algorithm that iteratively matches the input parameter for the anomalous dimension
@xmath116 with the output parameter @xmath117 to a desired precision .
the algorithm begins with a ( reasonable ) guess for @xmath116 and calculates the corresponding @xmath120 which will be different from fig .
[ criticalsol ] .
then it computes @xmath117 associated with the critical potential , and finally iterates by choosing a new value of @xmath116 closer to @xmath117 .
further details of the algorithms are discussed in appendix [ appendix_matching_algorithm ] .
the result of the iteration is given with great accuracy in subsect .
[ subsection_ising3d ] .
once both a critical value @xmath121 of @xmath90 and the corresponding solution @xmath122 are identified , it is possible to compute the critical properties of the solution in the form of critical exponents .
these are obtained by linearizing in the deviations from the critical solution , which means inserting the deformed solution @xmath123 into and linearizing in the infinitesimal expansion parameter @xmath124 . here
@xmath125 is a reference scale , and @xmath126 plays the role of the critical exponent of the deformation .
the linear order in @xmath124 gives an ode for the fluctuation @xmath127 that in canonical form reads @xmath128 ^ 2\,\times \\ & \bigl ( ( \theta - d)\delta v(\varphi ) + \frac{1}{2}(d-2+\eta)\varphi\,\delta
v'(\varphi ) \bigr)\ , .
\end{split}\ ] ] for later use , it proves convenient to perform a transformation that redefines the deformation as @xmath129 ^ 2\right)\,\delta v(\varphi),\\ & \alpha=\frac{d+2}{4c_d}\frac{d-2+\eta}{d+2-\eta}>0\ , , \end{split}\ ] ] such that the linearized flow equation does not contain the first derivative @xmath130 .
the explicit form of the transformed ode for @xmath131 is involved and we found it not particularly instructive ; thus we do not write it down explicitly .
the second derivative @xmath132 in the linear fluctuation equation is an even function and we may assume that the deformations @xmath127 are even or odd functions which means that they have the same parity as @xmath133 ( parity - even ) or the opposite parity ( parity - odd ) . again , due to the linearity of we are free to choose the normalization of @xmath127 at will .
this leaves us , modulo the global normalization , with two possible choices for the boundary conditions of the fluctuations at @xmath74 @xmath134 for parity - even and parity - odd deformations , respectively .
the same is of course true for the transformed deformations @xmath131 .
the spectrum of the critical exponents @xmath126 of the deformations is quantized when the deformations are required to be polynomially bounded , thus adding a further condition at large values of the field .
we denote the quantized spectrum of exponents as @xmath135 , where @xmath136 label the even / odd parity of the deformation and the index @xmath137 orders the exponents starting with the fluctuations with the highest positive critical exponent downward . for the actual determination of the spectrum it is important to realize that , in our setup , the solution @xmath133 is determined numerically and exists within a finite range only , @xmath138 .
hence it is necessary to replace the boundary condition at infinity with appropriately chosen conditions at @xmath139 to reproduce the quantized spectrum of fluctuations .
there are two natural ways to do so .
on the one hand the requirement that the fluctuations are subleading at infinity can be translated , within numerical accuracy , to the requirement the fluctuations must tend to zero at the actual numerical boundary @xmath140 .
this can be achieved numerically with the help of the shooting method .
the errors of introducing artificial boundary conditions for @xmath127 can be controlled when one relaxes the condition by setting @xmath141 , where @xmath142 is an arbitrary numerical constant .
typically , the critical exponents show a very mild dependence ( of order @xmath143 ) on @xmath142 even on the very large range @xmath144 $ ] .
this is due to the fact that the factor @xmath145 $ ] appearing in mitigates the effects of any boundary condition by suppressing it in the limit @xmath146 .
the dependence on the approximate boundary condition can be relaxed even further by considering the transformed fluctuation , for which it is natural to require @xmath147 . on the other hand ,
since the numerical solution exists only in a finite range , it seems natural to impose periodic boundary conditions for the even fluctuations and antiperiodic for the odd ones .
we studied both cases by implementing the very efficient slac derivative method for the transformed fluctuations defined in to compute the critical exponents . in this case , the potential in the range @xmath148 is discretized over a equidistant grid such that the schrdinger - type differential equation for @xmath131 turns into a ( large ) matrix - equation in @xmath7 space .
most importantly , the derivative with respect to @xmath7 is discretized as non - local slac derivative . in the past
this _ pseudo - spectral method _ based on trigonometric functions proved to be very efficient to accurately calculate the low - lying energies of schrdinger - type operators .
one may also estimate the characteristic width of an eigenmode @xmath131 , which can be compared with the maximal field @xmath139 to decide whether the solution can be trusted and to what extent the spectrum is affected by finite size effects .
furthermore , the convergence of the spectrum can be checked by increasing the number of grid points .
not unexpectedly , the critical exponents obtained with the spectral method yield the most accurate results . we refer to appendix [ appendix_slac ] for more details on the method . in this subsection
we provide tabulated results for the critical solution of in three dimensions that corresponds to the ising universality class . in each case
we studied the critical value @xmath107 with at least 10 digits precision , while the anomalous dimensions were determined with at least seven digits @xmath149 the critical exponents were always determined with at least six digits precision . in table
[ table_ising3d ] the first three rows correspond to the lpa schemes described in the previous sections , including the lpa for which the input anomalous dimension @xmath150 , the lpa obtained by matching @xmath116 with the anomalous dimension at the minimum using , and the lpa obtained by matching @xmath116 with the anomalous dimension at the minimum using .
the fourth row gives the result from a lpa in which the best known value of the anomalous dimension is taken from @xcite as an input parameter , while the last three rows contain the best known results from various techniques @xcite , from lattice s finite - size scaling @xcite , and from the conformal bootstrap approach @xcite .
we provide values for the critical parameter @xmath107 , for both input and output anomalous dimensions , as well as for the critical exponent @xmath151 governing the scaling of the correlation length which is defined as the inverse of the largest even critical exponent @xmath152 .
we also give numerical estimations for the dimensionless coupling , defined as @xmath153 note that the definition refers to renormalized quantities and bears an intrinsic local potential scheme dependence .
the comparison of this critical coupling with analog definitions coming from a lattice s bare action ( see for example @xcite and references therein ) might not be straightforward . the explicit relation between and its lattice s counterpart could perhaps be computed along the lines of a non - perturbative generalization of @xcite .
.critical parameters for the ising class in @xmath100 .
the parameters are explained in the main text .
the shooting and slac - method yield the same values for @xmath151 . [ cols="<,>,>,>,>,>",options="header " , ] as for the supersymmetric extension of the ising class in subsect .
[ subsection_susy_ising2d ] , two critical exponents can already be predicted with the help of the results of appendix [ appendix1 ] . in the present case
the superpotential has the opposite parity as for the ising class and both deformations discussed in the appendix appear as odd deformations of both the potential and the superpotential .
the new relations in fact determine the first two exponents in tab .
[ table_susy_triising2d_odd_exponents ] and take the values @xmath154 which hold true within our numerical precision .
the first relation is the equivalent of the superscaling relation ; however , here it does not determine the first critical exponent of the even deformations and thus does not give a direct relation between @xmath151 and @xmath35 .
the two flow equations and admit an analytic continuation to arbitrary fractional dimensions @xmath155 , leading to the question of whether it is possible to generalize the multicritical models to spaces with fractal properties .
this possibility has been investigated at length for the simple scalar model in @xcite , whose results we partly reproduce below for completeness .
however , the same is not true for the continuation of the wz - model with one supercharge .
one important physical motivation to continue
the system to fractional dimensions is that it was observed in @xcite that there is a correspondence between the scalar models in fractional dimensions with local ( short - range ) interactions , and the same models in two or three dimensions with non - local ( long - range ) interactions .
it is thus an intriguing possibility that the analytic continuation of our wess - zumino model to fractional dimensions plays a role in describing long - range effects in the phenomenon of emergent @xmath156 supersymmetry in topological superfluid phases @xcite .
the interesting question to answer now is the following : what are the critical dimensions at which new critical models appear when continuously following @xmath8 from @xmath2 ( where very probably there is only the gaussian model ) to @xmath1 ( where there are infinitely many models ) ?
the analytic continuations of the wz - model with more than one supersymmetry had already been studied in the literature in @xcite with the purpose of interpolating between the @xmath157- and @xmath158-theorem , and in @xcite where the conformal bootstrap of the model was considered .
the corresponding models all contain ( on - shell ) two real scalar fields , have two supersymmetries in @xmath159 dimensions and relate to superconformal @xmath160 models in @xmath1 dimensions .
this family of models is different from those considered in the present work .
our models have ( on - shell ) one scalar field and one supersymmetry in @xmath159 dimensions and relate to superconformal @xmath5 models in @xmath1 dimensions . motivated by both the landau - ginsburg theory and the results of the local potential approximation
, we assume that the local operators in the effective potential are the main actors when it comes to developing a new criticality class .
a new ( upper ) critical dimension appears every time a new local operator becomes marginal . to show this ,
let us first concentrate on the scalar field model and equation .
the set of simple local operators that satisfy the boundary conditions is exhausted by the powers @xmath161 for @xmath162 .
each of these local operators becomes marginal when the corresponding coupling is canonically dimensionless ; the dimensionality at which this happens is thus critical . in the language of the most appropriate way to find the critical dimension
is by observing that an operator is marginal if @xmath163\right|_{v(\varphi)=\varphi^{2n}}=0\,,\ ] ] which can be explicitly solved for @xmath8 to obtain the critical dimensions @xmath164 the series of critical dimensions accumulates at @xmath165 for @xmath166 when @xmath101 .
the plots of the end points @xmath91 of the integration of for the first few critical dimensions are depicted in fig .
[ plot_critical_dimensions_scalar ] .
each curve can be deformed continuously into the others by changing @xmath8 in .
it is easy to see that , by lowering the dimension , new singularities appear in the plot corresponding to new critical models .
the way in which new singularities appear is as follows : let us start from the upper critical dimension @xmath167 of the @xmath64 model where a new singularity appears to the left of @xmath88 .
it corresponds to the wilson - fisher fixed point of the ising class which we already studied in subsect .
[ subsection_ising3d ] . for @xmath168 with infinitesimal @xmath124
this singularity is arbitrarily close to the gaussian fixed point solution that is generally considered within perturbation theory .
if we follow @xmath8 further down to @xmath169 the singularity has moved left to a finite value of @xmath90 and just below @xmath170 a new singularity appears to the right of @xmath88 corresponding to the tricritical ising model .
again , this singularity can be followed for decreasing values of @xmath8 while it moves further to the right .
the picture that we just described can be iterated further : each time a new critical dimension is crossed a new singularity emanates from @xmath88 .
the singularities appear with an alternating sign of @xmath171 , and this explains the ordering of the models that we encountered , for example , in fig .
[ spike2dplot ] . for the first few critical dimensions @xmath172 .
the plot was obtained using the value @xmath173 .
, width=302 ] in each interval @xmath174 a new operator @xmath161 enters from its upper critical dimension defined through ; thus , it is able to satisfy the condition , which in turn implies a new branching of fig .
[ plot_critical_dimensions_scalar ] through the condition .
now we are ready to extend the above discussion to the supersymmetric models . recalling the form of the superpotential in
, a new critical dimension is expected to occur when a new operator @xmath175 becomes marginal , for which the symmetry of the model requires that @xmath176 . using the notation introduced in
, we thus seek the dimensionality for which @xmath177\right|_{w(\varphi)=\varphi^{m+1}}=0\,,\ ] ] which can be solved for @xmath8 to obtain the new critical dimensions @xmath178 again , the critical dimensions define a series which accumulate at @xmath165 for @xmath179 when @xmath101 .
the two sets of critical dimensions and strictly coincide for @xmath101 , while the difference is related to the fact that the on - shell superpotential defined in is subject to a multiplicative wavefunction renormalization which has no analogue in the scalar field theory .
now we can follow the solution of with boundary conditions or as a function of the dimension when lowering @xmath8 from @xmath180 to @xmath181 .
the result is depicted in fig.s [ plot_critical_dimensions_susy1 ] and [ plot_critical_dimensions_susy2 ] which show the function @xmath71 for even and odd boundary conditions for the superpotential .
for the first few critical dimensions @xmath182 .
the plot was obtained using the value @xmath173 .
, width=302 ] for the first few critical dimensions @xmath182 .
the plot was obtained using the value @xmath173 .
, width=302 ] we see that the dimensions @xmath182 for @xmath183 _ odd _ are critical for the plot in fig .
[ plot_critical_dimensions_susy1 ] , while those for @xmath183 _ even _ are critical for the plot in fig .
[ plot_critical_dimensions_susy2 ] . in both cases , by lowering @xmath8 a new singularity is produced in the plot when a new critical dimension is crossed .
the alternation of the singularities between the fig.s [ plot_critical_dimensions_susy1 ] and [ plot_critical_dimensions_susy2 ] is analogous to the one that occurs between the left and right sides of fig .
[ plot_critical_dimensions_scalar ] and was expected on the basis of the relation between the parameters @xmath90 and @xmath184 . since in our approximation
only the operators of the form @xmath185 can determine the occurrence of new critical dimensions , in the limit of zero anomalous dimension our result coincides with the set of critical dimensions given in @xcite where a simple yukawa model is considered .
for this reason , it is tempting to speculate that the yukawa and supersymmetric universality classes are the same .
this , however , is true only to some extent , as the allowed deformations of the wess - zumino model are constrained by supersymmetry if compared to those of a yukawa system , therefore changing the spectrum of the theory ( see @xcite for comments on this model from the point of view of the conformal bootstrap ) .
we studied the functional renormalization group flow of simple scalar field theories and their @xmath5 supersymmetric extensions in various local potential approximation schemes using a scaling solutions approach , pioneered in @xcite , in order to determine the stationary solutions of the flow .
our approach to the scalar model extends the work in @xcite , while the results for the supersymmetric wess - zumino model are new .
both families of models exhibit the expected universality classes in terms of stationary solutions of the flow . in three dimensions
we confirm the presence of both the ising class and its supersymmetric counterpart , while in two dimensions infinitely many stationary solutions can be observed which we identify with the minimal models of cft and super - cft .
it is rather exceptional that the infinite tower of minimal models is visible within such a simple truncation of the renormalization group flow that contains only a local potential and an anomalous dimension for the field . at various stages
, we offered a new understanding of the mechanism with which new critical solutions appear as stationary solutions of the flow of the local potential . the key idea is tightly connected to ( and best visualized through ) the ginzburg - landau theory : new critical models appear whenever new local operators turn into relevant deformations of the spectrum of the theory .
we explicitly observe that the appearance of each model is governed by the canonical scaling of a specific operator which can be pinpointed , for example , by analytically continuing the renormalization group flow between two and three dimensions .
the critical models can be identified as discontinuities in the plot of the domain of existence of all possible models using both scaling arguments and the landau description that counts the number of nodes of the solutions @xcite . for the sake of the exposition
, we concentrated our attention on the ising and tricritical ising classes , and their @xmath5 supersymmetric counterparts .
most importantly , we introduced two new methods to compute the relevant deformations of these models : the shooting and the pseudo - spectral slac methods .
while the shooting method has already been applied in @xcite to compute only the critical exponent of the correlation length , we showed that it can be used quite effectively to obtain many low - lying exponents .
one important ingredient of the present work is the application of the slac derivative method to the computation of the spectrum .
this latter approach can in fact easily produce a large number of critical exponents with one single calculation , effectively beating all other computationally slower algorithms and showing a mild dependence on the boundary conditions .
the only limitation of the slac derivative method is in the size of the numerically determined critical solution , which is finite in our implementation .
we thus expect the method to work even better when global solutions are available , such as in @xcite .
the functional renormalization group method is an efficient tool to analytically continue the flow in parameters that are fixed in most alternative approaches , for example lattice simulations , such as the dimensionality of the system .
we thus provided an analytic continuation of the @xmath5 wess - zumino model between two and three dimensions that , to the best of our knowledge , is new .
this continuation can continuously follow the stationary solutions of the local potential flow from three to two dimensions , showing explicitly what the critical dimensions are at which new models appear . without wave function renormalization
the critical dimensions of the supersymmetric wess - zumino model agree with those of the yukawa model with the same number of on - shell degrees of freedom calculated in @xcite .
it is an interesting open question whether this remains true beyond the lpa truncation and , in particular , with non - zero anomalous dimensions .
a particular feature of our work is that we can give a self - consistent description of any critical model within a very simple local potential approximation of the renormalization group flow .
we provided an extensive and detailed numerical analysis of ising and tricritical ising models .
our numerical results are in good agreement with those obtained by dedicated monte carlo simulations and by cft for the critical exponent of the correlation length @xmath151 , but typically fail in determining the anomalous dimension @xmath35 accurately .
a better determination of the anomalous dimension within the functional approach is crucial for improving the accuracy of the method .
it is natural to compare our results for the simple scalar field theory with morris contributions to the derivative expansion of the renormalization group @xcite , in which the local potential approximation is complemented by a field - dependent wavefunction renormalization and for which an additional scaling symmetry ( due to the choice of a powerlaw cutoff ) allows to determine the anomalous dimension from the boundary conditions of the flow .
we find a qualitative agreement with the results of @xcite for both the ising and tricritical ising classes .
as expected , the agreement becomes increasingly worse when the field dependent wavefunction renormalization of @xcite deviates from the value one .
the lesson to be learned here is that , while the local potential approximation works extraordinarily well in showing that the critical models are visible in our functional renormalization group approach , better numerical estimates of the critical parameter can only be achieved by enlarging the local potential truncation to include further operators .
therefore , an extension of our analysis to truncations of the effective action which account for a larger space of operators in functional form , such as a field dependent wavefunction renormalization , represents the most important and compelling prospect of our work .
another natural comparison for our work would be the recent developments on the conformal bootstrap approach @xcite .
the latter uses ward - identities of conformal field theory to constrain the operator algebra and provide accurate predictions for the critical exponents of the ising model . in our approach
, however , conformal invariance is an outcome of the requirement of scale invariance by spotting stationary solutions of the renormalization group flow and applying the landau classification to the solutions .
thus it is tempting to speculate that in a functional renormalization group approach it might not be necessary to use the full power of conformal field theory , but simply enhance the correlators appropriately to correctly capture the long range correlations in the vicinity of a critical point .
such an extension of the functional truncation might roughly follow the lines of @xcite , or be dictated by the desire of maintaining two - loops universality when continuing the results to four dimensions @xcite ( see also @xcite for a conformal field theory perspective ) .
[ [ acknowledgments . ] ] acknowledgments .
+ + + + + + + + + + + + + + + + we are grateful to t. r. morris and l. zambelli for valuable comments on an earlier draft of the paper .
the research was supported by the deutsche forschungsgemeinschaft ( dfg ) graduate school grk 1523/2 .
a. wipf thanks the dfg for supporting this work under grant no .
wi 777/11 - 1 .
o. zanusso thanks the dfg for supporting this work under grants no .
gi 328/7 - 1 and gi 328/6 - 2 ( for 723 ) .
let us consider a general renormalization group evolution equation for a scale dependent `` potential '' @xmath186 of the form @xmath187+{\cal f}_{h}[h'']\ , , \\ & { \cal s}_{h}[h , h';\varphi]\equiv - d_h h(\varphi)+ d_\varphi\ , \varphi\ , h'(\varphi)\ , . \end{split}\ ] ] as done for we introduced a scaling part @xmath188 $ ] that contains some general ( possibly almost canonical ) dimensions @xmath189 and @xmath190 for the dimensionful counterparts of the field @xmath7 and the potential itself respectively .
we assume that the non - trivial part of the flow @xmath191 $ ] only depends on @xmath192 and is otherwise arbitrary .
let us also introduce a @xmath17-stationary solution @xmath193 of of given parity .
the linearized equation for the fluctuations @xmath194 defined by @xmath195 takes the form @xmath196\delta h''(\varphi)=0\,,\ ] ] in which the prime denotes the derivative with respect to the argument .
for any @xmath193 with fixed parity there always exist two fluctuation modes with fixed parity and critical exponents that are directly related to @xmath189 and @xmath190 .
the first one is the magnetic deformation @xmath197 .
it is easy to see by direct substitution that this deformation is normalized at the origin by @xmath198 and solves for the critical exponent @xmath199 .
the other fluctuation is @xmath200 . to see this and to determine the corresponding critical exponent one computes @xmath201+{\cal f}_{h}[h^{*\prime\prime}]\bigr)=0\ , , \end{split}\ ] ] and compares the resulting expression with the fluctuation equation .
the comparison reveals that the critical exponent of this second deformation is @xmath202 .
the deformation has opposite parity compared to the fixed point solution @xmath193 .
in the specific example of the scalar field theory these two deformations have critical exponents @xmath203 the ratio of the second over the first exponent is the well known thermodynamical exponent @xmath204 .
the exponents always show up in our numerical computations .
since in the scalar case the fixed point solution @xmath205 is even , both exponents are contained in the spectrum corresponding to the odd fluctuations .
given the simple relation of these exponents with the anomalous dimension they can be used , in principle , to test the quality of the numerical algorithm used when studying the fluctuation spectrum .
we describe here the matching algorithm that was applied for obtaining the lpa solutions in greater detail .
similarly as in the introduction , we do this for the three dimensional ising class , but the considerations will apply straightforwardly to all other examples . as discussed in the main text ,
the stationary solutions of equation depend parametrically on the anomalous dimension and the required input value when solving this equation was denoted by @xmath116 . for each value of the parameter @xmath116
we can compute a corresponding value of @xmath107 by localizing the appropriate singularity in the plot of the value @xmath91 .
for reasonably small values of @xmath206 the plot of @xmath91 is qualitatively similar to fig .
[ spike3dplot ] . in fig .
[ matching3dplot ] we plot as a solid line the curve parametrizing the location of @xmath107 as a function of @xmath116 . for our definition of lpa scheme ,
we need the anomalous dimension computed at the minimum @xmath37 of the solution , which we dubbed @xmath117 , to coincide with @xmath116 .
this is not always the case , as shown in fig .
[ matching3dplot ] where the value of @xmath117 is plotted as a dashed curve .
the mapping @xmath207 is only defined locally , as its definition uses information about the potential at its ( absolute ) minimum , which is not always reached by the numerical solution .
the intersection of the two curves gives , by definition , the condition @xmath208 , and thus a critical solution in the sense of the lpa. in the lpa of the 3@xmath8 ising model through the matching condition @xmath209 .
the solid curve parametrizes the location of @xmath210 as a function of @xmath116 , while the dashed curve is a plot of @xmath117 of the solution corresponding to @xmath107 for the given @xmath116 .
the gray dashed curve represents the anomalous dimension of solutions for which @xmath209 , but it is computed by integrating non critical values of @xmath90 .
the dot gives , by construction , a solution which is both critical and satisfies the matching condition.,width=264 ] the scales of fig .
[ matching3dplot ] are particularly instructive , as they make clear that any numerical error in @xmath211 induces an error of @xmath35 which is at least one order of magnitude bigger . among the systems studied in this paper ,
the @xmath212 ising model is the one that shows the smallest error propagation from @xmath107 to @xmath35 . in other cases
the amplification is by two or three order of magnitudes .
since @xmath107 is the most important parameter as it controls the precision with which all other parameters are determined , we decided to determine it with precision of about @xmath213 or higher .
the anomalous dimension @xmath35 is then typically determined at a precision in the range @xmath214 depending on the model considered .
the newton - type algorithm for the matching works as follows .
once a sufficiently accurate trial value for @xmath215(lpa ) is found ( for example with precision @xmath143 as seen in fig .
[ matching3dplot ] ) , this value is used to determine @xmath107 and subsequently @xmath117 .
if @xmath116 and @xmath117 do not match to the desired precision , the value of @xmath116 is changed in the direction of @xmath117 by a step smaller than the desired precision and the procedure is then iterated .
the step size should be smaller than the desired precision ; otherwise , occasionally the algorithm might get stuck in loops , which are caused by the fact that the two curves of fig .
[ matching3dplot ] are almost parallel if plotted with the same scale on both axes .
this very simple algorithm yields precise estimates for @xmath107 .
when one tries to calculate the fluctuation spectrum of a given fixed point solution one meets the problem that the solution of the fixed point equation is only known up to a maximal value @xmath216 of the field ( cf . the corresponding discussion in subsect . [ subsection_scaling_solutions ] ) .
this is due to discretization and rounding errors involved in the numerical computation of the critical values of the initial conditions . even for initial conditions extremely close to criticality one finds @xmath217 .
both the shooting and spectral - slac methods have been adapted for ( various ) boundary conditions at the maximal value of the field @xmath218 for stationary solutions of equation parametrized by @xmath90 to extract the eigenfunction and eigenvalues of the corresponding fluctuation operator .
this can be done for non - critical solutions and the critical solution . in fig .
[ sigma_dependence_plot ] we plot the exponents corresponding to the parity even deformations of the solutions of for @xmath100 with boundary conditions .
the plot includes the exponent @xmath219 of the constant deformation that corresponds to the scaling of the volume operator .
it was never listed in the tables since for all critical models it is just the dimension of the system . of the critical exponents @xmath220 for @xmath221 from top to bottom of the three dimensional ising class .
this dependence is computed using the shooting algorithm .
, width=302 ] the plot for the ising class in fig .
[ sigma_dependence_plot ] reveals many interesting features . all exponents @xmath222 seem to be continuous functions of the parameter @xmath90 with cusps at the critical values @xmath210 corresponding to the non - trivial fixed point and at @xmath88 corresponding to the gaussian fixed point .
the critical values of @xmath90 are marked with vertical dashed lines in the plot .
the critical exponents of all deformations are negatively divergent at @xmath86 and gradually increase with increasing curvature @xmath90 to their value at the ising class . beyond @xmath107
they continuously interpolate with the spectrum of the free theory at the gaussian point , where they attain their maximal values , and then their values decrease again .
each eigenvalue shows a distinct behavior near the critical points .
also note that the scaling exponent of the volume operator @xmath219 is not equal to @xmath159 below @xmath107 , but takes this value at @xmath107 and stays there up to @xmath88 . beyond this point it decreases again
albeit very slowly .
all higher exponents approach @xmath107 with a slope that differs between the left side and the right side . as is expected from the discontinuous change in the domain of existence of the potential ( see also fig .
[ discontinuity ] which shows almost critical solutions from the left and the right of the critical point ) .
we describe here the spectral - slac method for the computation of the critical exponents that was applied in the present work .
again we use the scalar field theory of sect .
[ section_simple_scalar ] as an example , but all considerations can be extended to the supersymmetric model of sect .
[ section_wz_model ] .
the stability analysis around the solution of the second order differential equation yields the linear second order differential equation that depends parametrically on the critical exponents @xmath126 .
the critical exponents @xmath126 must thus , in general , be fixed by the boundary conditions .
if a global solution to is present , the physically meaningful condition needed is that the fluctuations are polynomially bounded in the limit of large field @xcite , which in turn implies a quantized spectrum of critical exponents with an upper bound @xmath223 . our numerical solutions , however , do not extend to the infinite field ( see for example fig .
[ criticalsol ] ) ; therefore , we have to impose boundary conditions at the extremum @xmath71 of the domain of validity , rather than at infinity .
the influence of these auxiliary boundary conditions can be minimized by assuming periodicity , in which case the amplitude of the fluctuations can vary freely near @xmath71 rather than being zero as enforced by the shooting method . with these premises , there is a very fast , precise and efficient algorithm to compute the spectrum of the critical exponents @xmath224 which is known the spectral - slac method , because it uses the lattice s slac derivative @xcite .
let the linear ode describing some fluctuations @xmath131 around the critical solution @xmath133 be of the form @xmath225 where we used the fact that any linear second order ode can be cast in a form without first order derivative ( for the scalar model this transformation is given in ) .
the functions @xmath226 and @xmath227 generally depend on @xmath133 : let us assume that the function @xmath226 is positive .
eqn . can be interpreted as the quantum mechanical problem of diagonalizing the hamiltonian @xmath228 with eigenvalues @xmath229 .
now we discretize the interval @xmath230 $ ] on which @xmath7 is defined by @xmath4 lattice points @xmath231 with constant separation @xmath232 .
the momentum takes its values in the dual lattice @xmath233 the lattice s slac derivative is defined as @xmath234 where @xmath235 is a complex - valued function on @xmath26 and @xmath236 is its fourier transform . differently from many standard lattice
s derivatives , the non - local slac derivative is given by a matrix whose entries are , in general , non - zero for any pair of lattice - points @xmath237 and @xmath238 . with respect to periodic boundary conditions the second derivative based on the slac - derivative reads @xmath239 with an odd number @xmath4 of lattice points . for anti - periodic boundary conditions ( suitable for the parity - odd fluctuations )
one uses the same matrix but an even number @xmath4 of lattice points .
this matrix is used to discretize in the form @xmath240 the discretized operator @xmath241 is an @xmath242 matrix whose eigenvalues can be computed using standard numerical methods . up to a sign
these are the exponents @xmath224 .
the method yields very accurate eigenvalues if the corresponding eigenfunctions have their main support within the domain @xmath230 $ ] and if the number of lattice points @xmath4 is sufficiently large to resolve the typical variations of the eigenfunction .
the biggest advantage of the slac method is the efficient suppression of discretization errors and ( for sufficiently large @xmath243 and @xmath4 ) the very accurate results for the low lying @xmath244 eigenvalues of the spectral problem with just one diagonalization .
the accuracy can be estimated by varying both the number of lattice points and the discretization interval @xmath230 $ ] .
the critical supersymmetric solutions obtained in subsections [ subsection_susy_ising3d ] , [ subsection_susy_ising2d ] and [ subsection_susy_tri_ising2d ] have a limited domain of existence , but can be extended to global ones by integrating through the point @xmath71 defined by @xmath245 in and . when approaching the critical point the numerator of , that represents the scaling of @xmath246 , fulfills @xmath247 thus balancing the corresponding zero of the denominator and giving rise to a finite @xmath248 in the limit .
our strategy for numerically integrating beyond @xmath71 is to perform a linear taylor expansion of @xmath249 and @xmath250 for which @xmath251 and @xmath252 change sign _ simultaneously _ at the critical point .
this can be achieved by choosing the almost critical initial condition @xmath253 , for which the solution will terminate when @xmath254 , while @xmath251 is still finite .
we illustrate in fig .
[ sprimeandfprime ] the behavior of the functions @xmath255 and @xmath251 in the vicinity of @xmath71 for the two dimensional supersymmetric ising class found in subsect .
[ subsection_susy_ising2d ] within the lpa scheme .
the plot shows that the two functions almost reach zero at the same point , before deviating because of the numerical uncertainty in the determination of the critical value of @xmath257 . the necessary condition for the existence of a finite and non
- zero @xmath258 is that the two functions intersect the @xmath7-axis at the same point @xmath243 .
thus , the crucial point is to accurately identify a small intervall in fig .
[ sprimeandfprime ] where @xmath259 and @xmath251 are almost zero but still depend linearly on @xmath7 .
then we perform a linear interpolation that ensures that both functions change sign and check that this happens at the same point @xmath243 .
once the linear interpolation beyond @xmath71 is made , the numerical integration of yields the solution for all values of the field . for large fields
the numerator and denominator slowly approach zero thus realizing the predicted scaling limit in . in fig .
[ globalsol2dsusy ] we show the plot of the on - shell effective potential extending beyond @xmath71 for the lpa supersymmetric ising class of subsect . [ subsection_susy_ising2d ] .
the extended solution does not encounter further singularities beyond @xmath71 even though its asymptotic behavior is governed by , which is very similar to the asymptotic behavior of scalar field theory in .
the natural question to ask at this point is th following : does the extension of the critical solution affect the spectrum of its fluctuations ?
it is possible to show , in general , that this is not the case .
let us consider the linearized equation for the fluctuations @xmath260 of @xmath249 around the critical solution , which is obtained by deforming @xmath261 in the renormalization group flow @xmath262 which can be easily obtained by applying a @xmath7 derivative to both sides of .
the linearized ode is of the form @xmath263 in which we introduced the functions @xmath264 which have to be evaluated at the critical solution @xmath265 . as we did in for the simple scalar model
, we transform the fluctuations in so that the transformed equation does not contain first derivatives with respect to the field .
this is done by defining the new fluctuations @xmath131 such that @xmath266 where we can choose @xmath267 or @xmath268 for the lower integration limit . in both cases , the transformed fluctuations solve a schrdinger - type equation of the form @xmath269 since @xmath270 and @xmath271 the transformation is singular at the point @xmath243 .
since in addition @xmath272 has positive slope at @xmath243 it follows that the exponential factor in vanishes when @xmath7 approaches @xmath243 from below and from above .
this means that @xmath273 obeys dirichlet boundary conditions at @xmath274 .
it follows that the spectral problem has two disjoint sets of solutions , solutions with @xmath275 and solutions with @xmath276 .
the only admitted fluctuations satisfy either the boundary conditions or , which are possible only for the transformed fluctuations for which @xmath268 , thus implying that @xmath277 is the most natural choice .
we explored numerically the spectrum of the global solution plotted in fig .
[ globalsol2dsusy ] using the slac derivative method discussed in appendix [ appendix_slac ] applied to the eigenvalue problem .
it can be checked that the spectrum does not receive any correction from the extension of the domain of existence of the fixed - point solution beyond @xmath243 .
j. zinn - justin , int .
phys . * 113 * , 1 ( 2002 ) .
a. pelissetto and e. vicari , phys .
* 368 * , 549 ( 2002 ) [ cond - mat/0012164 ] . c. wetterich , phys .
b * 301 * , 90 ( 1993 ) .
t. r. morris , int .
j. mod .
phys . a * 9 * , 2411 ( 1994 ) [ hep - ph/9308265 ] .
j. berges , n. tetradis and c. wetterich , phys .
rept . * 363 * , 223 ( 2002 ) [ hep - ph/0005122 ] .
t. r. morris , prog .
suppl . * 131 * , 395 ( 1998 ) [ hep - th/9802039 ] .
o. j. rosten , phys .
rept . * 511 * , 177 ( 2012 ) [ arxiv:1003.1366 [ hep - th ] ] .
d. f. litim and d. zappala , phys .
d * 83 * , 085009 ( 2011 ) [ arxiv:1009.1948 [ hep - th ] ] .
d. f. litim , jhep * 0111 * , 059 ( 2001 ) [ hep - th/0111159 ] .
d. f. litim , nucl .
b * 631 * , 128 ( 2002 ) [ hep - th/0203006 ] .
d. f. litim and l. vergara , phys .
b * 581 * , 263 ( 2004 ) [ hep - th/0310101 ] .
t. r. morris , jhep * 0507 * , 027 ( 2005 ) [ hep - th/0503161 ]
. j. f. nicoll , t. s. chang and h. e. stanley , phys .
lett . * 33 * , 540 ( 1974 ) .
c. bervillier , b. boisseau and h. giacomini , nucl .
b * 789 * , 525 ( 2008 ) [ arxiv:0706.0990 [ hep - th ] ] .
j. borchardt and b. knorr , phys .
d * 91 * , no .
10 , 105011 ( 2015 ) [ arxiv:1502.07511 [ hep - th ] ] .
a. codello , j. phys .
a * 45 * , 465006 ( 2012 ) [ arxiv:1204.3877 [ hep - th ] ] .
j. comellas and a. travesset , nucl .
b * 498 * , 539 ( 1997 ) [ hep - th/9701028 ] .
a. codello and g. dodorico , phys .
* 110 * , 141601 ( 2013 ) [ arxiv:1210.4037 [ hep - th ] ] .
i. boettcher , phys .
e * 91 * , no .
6 , 062112 ( 2015 ) [ arxiv:1503.07817 [ cond-mat.stat-mech ] ] .
r. percacci and g. p. vacca , phys .
d * 90 * , 107702 ( 2014 ) [ arxiv:1405.6622 [ hep - th ] ] .
a. codello , n. defenu and g. dodorico , phys .
d * 91 * , no .
10 , 105003 ( 2015 ) [ arxiv:1410.3308 [ hep - th ] ] .
t. r. morris , phys .
b * 334 * , 355 ( 1994 ) [ hep - th/9405190 ] .
t. r. morris , int .
j. mod .
b * 12 * , 1343 ( 1998 ) [ hep - th/9610012 ] .
t. r. morris , phys .
b * 329 * , 241 ( 1994 ) [ hep - ph/9403340 ] .
drell , m. weinstein and s. yankielowicz , phys .
d * 14 * , 487 ( 1976 ) l. canet , b. delamotte , d. mouhanna and j. vidal , phys .
b * 68 * , 064421 ( 2003 ) [ hep - th/0302227 ] .
m. mastaler , f. synatschke - czerwonka and a. wipf , phys .
nucl . * 43 * , 593 ( 2012 ) .
f. synatschke , h. gies and a. wipf , phys .
d * 80 * , 085007 ( 2009 ) [ arxiv:0907.4229 [ hep - th ] ]
. f. synatschke , h. gies and a. wipf , aip conf .
proc . * 1200 * , 1097 ( 2010 ) [ arxiv:0909.4189 [ hep - th ] ] .
f. synatschke , j. braun and a. wipf , phys .
d * 81 * , 125001 ( 2010 ) [ arxiv:1001.2399 [ hep - th ] ] .
m. heilmann , d. f. litim , f. synatschke - czerwonka and a. wipf , phys .
d * 86 * , 105006 ( 2012 ) [ arxiv:1208.5389 [ hep - th ] ] . c. bervillier , nucl .
b * 876 * , 587 ( 2013 ) [ arxiv:1307.3679 ] .
m. hasenbusch , phys .
b * 82 * , 174433 ( 2010 ) [ 1004.4486 ] .
d. simmons - duffin , jhep * 1506 * , 174 ( 2015 ) [ arxiv:1502.02033 [ hep - th ] ] .
p. bosetti , b. de palma and m. guagnelli , arxiv:1506.08587 [ hep - lat ] .
a. codello , m. demmel and o. zanusso , phys .
d * 90 * , no . 2 , 027701 ( 2014 ) [ arxiv:1310.7625 [ hep - th ] ] .
m. heilmann , t. hellwig , b. knorr , m. ansorg and a. wipf , jhep * 1502 * , 109 ( 2015 ) [ arxiv:1409.5650 [ hep - th ] ] . n. defenu , a. trombettoni and a. codello , arxiv:1409.8322 [ cond-mat.stat-mech ] . t. grover , d. n. sheng and a. vishwanath , science * 344 * , no . 6181 , 280 ( 2014 ) [ arxiv:1301.7449 [ cond-mat.str-el ] ] . s. giombi and i. r. klebanov , jhep * 1503 * , 117 ( 2015 ) [ arxiv:1409.1937 [ hep - th ] ] .
n. bobev , s. el - showk , d. mazac and m. f. paulos , arxiv:1502.04124 [ hep - th ] ; + n. bobev , s. el - showk , d. mazac and m. f. paulos , arxiv:1503.02081 [ hep - th ] .
g. p. vacca and l. zambelli , phys .
d * 91 * , no .
12 , 125003 ( 2015 ) [ arxiv:1503.09136 [ hep - th ] ] .
d. bashkirov , arxiv:1310.8255 [ hep - th ] .
s. el - showk , m. f. paulos , d. poland , s. rychkov , d. simmons - duffin and a. vichi , phys .
d * 86 * , 025022 ( 2012 ) [ arxiv:1203.6064 [ hep - th ] ] ; + s. el - showk , m. f. paulos , d. poland , s. rychkov , d. simmons - duffin and a. vichi , j. stat .
* 157 * , 869 ( 2014 ) [ arxiv:1403.4545 [ hep - th ] ] .
s. el - showk , m. paulos , d. poland , s. rychkov , d. simmons - duffin and a. vichi , phys .
lett . * 112 * , 141601 ( 2014 ) [ arxiv:1309.5089 [ hep - th ] ] .
a. codello and a. tonero , arxiv:1504.00225 [ hep - th ] .
s. rychkov and z. m. tan , j. phys .
a * 48 * , no .
29 , 29ft01 ( 2015 ) [ arxiv:1505.00963 [ hep - th ] ] . | we study the renormalization group flow of @xmath0-invariant supersymmetric and non - supersymmetric scalar models in the local potential approximation using functional renormalization group methods .
we focus our attention on the fixed points of the renormalization group flow of these models , which emerge as scaling solutions . in two dimensions
these solutions are interpreted as the minimal ( supersymmetric ) models of conformal field theory , while in three dimensions they are manifestations of the wilson - fisher universality class and its supersymmetric counterpart .
we also study the analytically continued flow in fractal dimensions between @xmath1 and @xmath2 and determine the critical dimensions for which irrelevant operators become relevant and change the universality class of the scaling solution .
we include novel analytic and numerical investigations of the properties that determine the occurrence of the scaling solutions within the method .
for each solution we offer new techniques to compute the spectrum of the deformations and obtain the corresponding critical exponents . |
one of the less - studied interactions within our galaxy concerns the effect of the passage of a globular cluster ( gc ) through the disk , on the disk itself . on average
, one globular cluster will cross the galactic disk about every million years . while the globular cluster is in the vicinity of the disk , its gravitational field will attract stars and gas in the disk . the result could be a density increase sufficient to cause star formation . whereas the impacts of high velocity clouds with the disk have been examined in some detail ( comern and torra , 1994 ) ,
as they cause gas compression and star formation ( and are perhaps the origin of the gould belt ) , the effect of gc impacts on the disk has received almost no attention .
the earliest mention of the effect of disk crossing by a globular cluster on the disk , is from brosche et al .
( 1991 ) who calculated the orbits of globular clusters ngc362 and ngc6218 , and suggested observers examine their latest crossing points , to see whether any remnants may be present .
wallin , higdon & staveley - smith ( 1996 ) developed the idea .
their main argument is that gravitational focussing of disk material , such as gas , towards the gc trajectory would occur .
they concluded that large ob star associations would form , possibly some 30 myr after the disk crossing , over a region of size @xmath010 pc .
their paper elicited little reaction , aside from two instances .
the first was by levy ( 2000 ) , who examined compression of disk gas by the shock wave produced during gc travel through the disk .
this paper appears never to have been cited , in spite of its importance to this question ; we refer to it below .
the second reaction to wallin et al s paper was the suggestion by r. rees and k. cudworth , ( reported in wright , 2004 ) that the star association ngc 6231 in the galactic disk is the result of such an interaction by globular cluster ngc6397 , less than 5myr ago .
they reached this conclusion using positional and proper motion data , together with a code to calculate orbits in the galactic potential .
they found the point of impact of ngc6397 to correspond to the position of open cluster ngc6231 , a finding we discuss later .
the phenomena that take place in disk crossings by gcs are akin to those encountered in minor mergers and galaxy harassment , where a large galaxy interacts with a smaller galaxy at low and high relative velocities , respectively .
cox et al . ( 2006 ) predict that the global star formation rate decreases with increasing mass ratio , so that mergers with mass ratio@xmath1 20 induce hardly any additional star formation in the more massive galaxy , above its quiescent level . in the case of gcs ,
the mass ratios are even larger , but relative velocities are lower and local effects in the disk may still be significant .
indeed , minor mergers have been proposed as the origin of nuclear starburst rings in some local galaxies ( e.g. mazzuca et al .
2006 ) .
the objective of this work is to examine whether , in principle , star formation , or at least disk material compression could occur in a gc impact , and to look for possible remnants in the galaxy . in section 2 , we describe an approach to modelling the interaction between a gc and the disk , which establishes how compression may occur via gravitational focussing .
this includes developing models of the galaxy and gc potentials to calculate the trajectories of gcs and test masses in the disk .
we then note an alternate mechanism for compression , namely shock wave generation .
if gas is present , such a mechanism will be more effective at inducing star formation . in section 3 , we apply the galactic models to locate the most recent impact areas of gcs in the disk . sections 4 and 5 use these location results to search for possible remnants in the galactic plane .
finally , section 6 provides conclusions .
a preliminary version of this work was reported in cropper & vande putte ( 2007 ) .
we examine first the case of a gas - free disk , and consider the effect of a gc crossing , using numerical simulations of a series of test masses representing disk material encountering a globular cluster as it crosses the disk .
these test masses are susceptible to the galactic potential , and to the potential of the globular cluster , but not to one another .
we later consider the case where gas is present at the impact site .
first , however , we use the next two subsections to describe the methods we employ to calculate orbits in galactic and gc potentials .
we use four galactic potentials for which straightforward analytical expressions are available .
we denote these da , fe , fl , and pa , as they appear in dauphole & colin ( 1995 ) , fellhauer et al .
( 2006 ) , flynn , sommer - larsen & christensen ( 1996 ) , and paczynski ( 1990 ) , respectively .
others have used these expressions in studies of galactic dynamics , for example kalirai et al .
( 2007 ) , law , johnston & majewski ( 2005 ) , pauli et al .
( 2003 ) , and dinescu , girard & van altena ( 1999 ) .
these potentials are all axisymmetric , so their expressions are simplest in cylindrical coordinates @xmath2 where @xmath3 is the distance to the z axis , @xmath4 is the azimuthal angle between the x axis and the projection of the position vector on the xy plane , and z the distance above the x , y plane . the analytical characteristics of the potentials appear in table 1 .
they consist of bulge , halo , and disk contributions described by different models , as follows .
v@xmath5 at 8kpc & _ _ m__@xmath6 at 100kpc + & & & & km s@xmath7 & 10@xmath8 @xmath9 + dauphole ( da ) & plummer & plummer & miyamoto - nagai & 225 & 7.9 + fellhauer ( fe ) & hernquist & log form & miyamoto - nagai & 224 & 9.3 + flynn ( fl ) & 2xplummer & log form & 3xmiyamoto - nagai & 222 & 11.8 + paczynski ( pa ) & miyamoto - nagai & log - atan form & miyamoto - nagai & 220 & 8.5 + the analytical expression for the potential from a plummer ( 1911 ) sphere is : @xmath10}^{1 / 2}}}}\ ] ] where @xmath11 the miyamoto - nagai ( 1975 ) potential s analytical expression is : @xmath12}^{2 } } \right\}}^{1 / 2}}}}\ ] ] the expression for the hernquist ( 1990 ) potential is : @xmath13 the logarithmic potential form is ( fellhauer et al .
2006 ) : @xmath14 the logarithmic - arctan combination potential is ( paczynski , 1990 ) : @xmath15}\ ] ] [ cols="^,^ " , ] the six areas and subtended fields of view from earth encompassed by the contours in figure 7 are smallest for ngc 3201 , ngc 6397 and ngc 6838 . as a result ,
these offer the best prospects for identifying remnants of crossings . for ngc 3201
, the area is enclosed within a radius of 7 on the sky , and a distance range of @xmath1622% from the nominal impact . for ngc 6397 , these radii are 3 , with distance @xmath1612% and for ngc 6838 , they are 6 , and @xmath1620% .
we searched the cds simbad database centred on regions of sky corresponding to these areas , and restricted the search to object types star cluster , cl * " , open galactic cluster , opc " , and association of stars , as * " as defined in the database .
this leaves 15 , 33 and 42 star groups respectively , for ngc 3201 , ngc 6387 , and ngc 6838 , and these are listed in tables 5 , 6 and 7 .
c 0820 - 360 & cl * & a , d & 1 + c 0815 - 369 & cl * & a & 2 + c 0813 - 306 & cl * & a , d & 1 + ngc 2546 & cl * & a , d & 1 + eso 430 - 18 & cl * & a & 1 + @xmath17kpr2005@xmath18 $ ] 45 & cl * & a & 3 + c 0812 - 362 & cl * & a , d & 1 + c 0810 - 324 & opc & a , d & 1 + c 0812 - 318 & opc & a , d & 4 + ngc 2588 & opc & a , d & 5 + + potential candidates are c 0805 - 322 , [ dbs2003 ] 16 , c 0807 - 343 , c 0809 - 318 , dsh j0807.1 - 3603 , pending distance and age data + refs : 1 , kharchenko , 2005a ; 2 , lindoff , 1968 ; 3 , kharchenko et al .
, 2005b ; 4 , giorgi et al .
, 2007 ; 5 , baume et al . , 2004 .
+ c 1715 - 382 & cl * & d & 1 + @xmath17bds2003@xmath18 $ ] 97 & cl * & d & 2 + @xmath17bds2003@xmath18 $ ] 98 & cl * & d & 2 + @xmath17bds2003@xmath18 $ ] 164 & cl * & d & 2 + @xmath17bds2003@xmath18 $ ] 99 & cl * & d & 2 + c 1714 - 355 & cl * & a , d & 4 + @xmath17dbs2003@xmath18 $ ] 179 & cl * & d & 5 + ngc6318 & opc & a & 3 + + potential candidates are [ dbs2003 ] 120 ; [ dbs2003 ] 121 ; [ dbs2003 ] 122 ; [ dbs2003 ] 123;mfsw vi ; mfsw v ; mfsw iv ; c 1717 358 ; mfsw iii ; mfsw ii ; mfsw i ; dsh j1715.7 - 3843 ; [ bdb2003]g351.61+.17 ; [ bds2003 ] 119 ; [ dbs2003 ] 118 ; c 1712 - 393 ; c 1727 - 370 ; cl vdbh 214 ; c 1720 - 378 ; c1721 - 389 ; c 1728 - 368 ; dsh j1713.2 - 3942 ; c 1715 - 387 ; [ dbs2003 ] 166 ; [ dbs2003 ] 165 , pending distance and age data + refs : 1 , piatti & clari , 2002 ; 2 , bica et al . , 2003 ; 3 , piatti et al .
, 2000 ; 4 , kharchenko et al . , 2005b ; 5 , borissova et al , 2005 + @xmath17kpr2005@xmath18 $ ] 104 & cl * & a , d & 1 + c 1915 + 194 & cl * & a & 2 + @xmath17bds2003@xmath18 $ ] 155 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 153 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 154 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 146 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 152 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 150 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 151 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 145 & cl * & d & 3 + ob cluster in w51 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 148 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 147 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 144 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 143 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 142 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 141 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 140 & cl * & d & 3 + ngc6793 & cl * & a , d & 1 + @xmath17bds2003@xmath18 $ ] 139 & cl * & d & 3 + @xmath17bds2003@xmath18 $ ] 138 & cl * & d & 3 + dsh j1930.2 + 1832 & opc & a d & 4 + dsh j1933.9 + 1831 & opc & a , d & 4 + ngc6802 & opc & a , d & 5 + c 1942 + 174 & opc & a , d & 1 + + potential candidates are [ bds2003 ] 12 ; [ bds2003 ] 156 ; c 1926 + 173 ; dsh j1926.0 + 1945 ; [ bds2003]157 ; [ bds2003 ] 13 ; c 1926 + 147 ; dsh j1942.3 + 1939 ; c 1916 + 156 ; dsh j1942.8 + 1530 ; c 1940 + 210 ; dsh j1912.0 + 1716 ; c 1922 + 136 ; dsh j1937.3 + 1841 ; cl czernik 40 ; dsh j1925.2 + 1356 ; cl alessi 57 , pending distance and age data + refs : 1 , kharchenko et al . , 2005b ; 2 , carraro et al .
, 2006 ; 3 , bica , et al . , 2003 ; 4 , kromberg et al . , 2006 ; 5 , netopil et al . , 2007 .
+ we searched the relevant literature for details of age and distance .
where this information is available , we noted for each star group in tables 5 , 6 , and 7 , the parameter that is incompatible , age ( a ) by a factor of at least two , and distance ( d ) outside the bounds given above .
we conclude that of a total of fifteen star groups that lie in the direction of ngc3201 s possible impact remnants , ten are definitely not remnants , on the grounds of incompatible age or distance ( table 5 ) . in the remaining five cases , it is possible that the groupings could result from the impact , pending further work on determining the properties ( age , distance ) of the associations . examining the corresponding data for ngc6397 ( table 6 ) , out of thirty - three candidates , eight can be excluded .
the corresponding data for ngc6838 ( table 7 ) indicates that twenty five out of forty two candidates , can be excluded .
this means that we are able to discount some gcs , and identify potential candidates , pending availability of further data .
level of ngc 6397 impacts ( fe potential ) , at the time of gc crossing .
the triangle and circle represent the points corresponding to today s distance estimate @xmath19 , respectively.,width=336 ] an alternative approach to linking star formation to gc disk crossings is to work in reverse , looking for evidence of unusual star formation , and asking whether this could be due to a gc disk crossing . for this
we have concentrated on super star clusters ( sscs ) .
portegies zwart ( 2004 ) identifies five such sscs : arches , quintuplet , ngc3603 , westerlund 1 , and westerlund 2 .
their distance from the galactic centre and age appear in table 8 .
we first look for progenitors amongst our sample of fifty - four gcs as we have already determined their crossings .
it is possible to eliminate some of these , on the basis of the time elapsed since the last crossing .
the time @xmath20 since a crossing is given by @xmath21 where @xmath22is the time interval between disk crossing and the onset of star formation , @xmath23 is the duration of star formation , and @xmath24 is the age of the cluster .
the first time ( @xmath22 ) is not well known , so three cases are considered here .
firstly , it is possible that this time is small compared to the age of the cluster , so that we can neglect it .
secondly , we adopt the value of 15myr given by lpine & duvert ( 1994 ) for the case of star formation after a high velocity cloud impacts the disk . finally , we take the value of 30myr suggested by wallin et al . (
for the star formation time , we adopt a value of @xmath25=2 @xmath2610@xmath27 years , suggested by mckee and tan ( 2002 ) , for massive stars ( @xmath28 ) .
taking the value @xmath22=30 myr , and an average ssc age of @xmath29=5 myr , we are able to identify 24 gcs as not being ssc progenitors , because the time since impact is too long compared with 35myr , using the fe potential . as a result
, the list of fifty - four gcs to consider for monte carlo analysis is reduced to 30 candidates . a monte carlo analysis of these 30 candidates , similar to those above , using the fe potential
, then shows further that the sscs could not have been produced by any of these 30 gcs , with the exception of ngc362 , 5897 , 6093 , 6316 , 6522 , 6584 , and 6779 , whose impact footprint is compatible with both the arches and quintuplet .
hence here too , some gcs can be discounted , while others may be potential progenitors .
having considered the fifty - four gcs , we then broaden our scope to all one hundred and fifty - five gcs in the harris catalogue , and note that we can also screen for candidates by using a weaker method but which does not require knowledge of any proper motions or radial velocities . as the position of the gc at the time of impact is the same as the position of the ssc s ancestral gas cloud , it is possible to calculate the minimum velocity @xmath30 required for the gc to cover the distance from that position , to its present - day position within a time @xmath31 .
the escape velocity at the galactocentric distances ( @xmath32 ) in table 8 can then be calculated from the potential , using the relation @xmath33 ( binney and tremaine , 1994 ) . if @xmath30@xmath34 @xmath35then the gc s last disk crossing can not be the event that triggered the formation of the ssc .
we made the comparison for all gcs in the harris catalogue , all five super clusters , the fe potential , and three times @xmath22 discussed above .
table 9 summarises the analysis .
we can identify the gcs that are definitely not progenitors of the sscs .
as expected and evident in table 9 , the case with the smallest set of excluded gcs is that where @xmath22=30 myr , and table 10 lists the definitely excluded gcs . on the other hand ,
if star formation as a result of disk crossing is prompt ( @xmath22@xmath00 ) then we can exclude _ all _ gc crossings as triggering the formation of both ngc3603 and westerlund 2 .
if this is generally the case , it may indicate that sscs are not generated by gc disk crossings .
@xmath32 ( kpc ) & age myr & refs .
+ arches & @xmath36 & @xmath37@xmath38 & 1 + quintuplet & @xmath39 & @xmath40@xmath41 & 2 + ngc3603 & 8@xmath42 & @xmath43 & 3 , 4 + westerlund1 & @xmath44 & 1 - 10 & 5 + westerlund2 & 9.8@xmath45 & @xmath46 & 6 + + refs : 1 , figer et al .
, 2002 ; 2 , figer , mclean , morris , 1999 ; 3 , stolte et al . , 2004 ; 4 , stolte et al . , 2006 ; 5 , vrba et al . , 2000 ; 6 , rauw et al . , 2007 .
+ @xmath22 = 15myr & @xmath22 = 30myr + arches & 20 & 76 & 93 + quintuplet & 34 & 80 & 93 + ngc 3603 & 0 & 16 & 75 + westerlund 1 & 1 & 60 & 82 + westerlund 2 & 0 & 17 & 73 + am1 & a , q , n , w1 , w2 & ngc5634 & n , w1 , w2 & ngc7492 & n , w1 , w2 + am4 & n , w1 , w2 & ngc5694 & a , q , n , w1 , w2 & pal1 & n , w2 + arp2 & n , w1 , w2 & ngc5824 & n , w1 , w2 & pal2 & a , q , n , w1 , w2 + eridanus & a , q , n , w1 , w2 & ngc6205 & w2 & pal3 & a , q , n , w1 , w2 + esosc06 & w1 & ngc6229 & a , q , n , w1 , w2 & pal4 & a , q , n , w1 , w2 + ic1257 & n , w1 , w2 & ngc6341 & n , w2 & pal5 & n , w1 , w2 + ic4419 & w1 & ngc6426 & n & pal11 & n , w2 + ngc1261 & w1 & ngc6715 & n , w1 , w2 & pal12 & n , w2 + ngc1904 & w2 & ngc6779 & n , w2 & pal13 & n , w1 , w2 + ngc2419 & a , q , n , w1 , w2 & ngc6864 & n , w2 & pal14 & a , q , n , w1 , w2 + ngc4147 & n , w1 , w2 & ngc6934 & n , w2 & pal15 & a , q , n , w1 , w2 + ngc5024 & n , w2 & ngc6981 & n , w2 & pyxis & a , q , n , w1 , w2 + ngc5053 & n , w2 & ngc7006 & a , q , n , w1 , w2 & rup106 & w1 + ngc5272 & w2 & ngc7078 & n , w2 & terzan7 & n , w1 , w2 + ngc5466 & n , w2 & ngc7089 & n , w2 & terzan8 & n , w1 , w2 + + a = arches ; q = quintuplet ; n = ngc3603 ; w1=westerlund1 ; w2=westerlund2 +
globular cluster interaction with a galactic disk has been studied in detail from the viewpoint of the gc , see for example gnedin , lee & ostriker ( 1999 ) . on the other hand ,
the effect on the disk has received scant attention ( brosche et al .
1991 , wallin et al .
1996 , levy 2000 ) .
the suggestions in these papers is that gc disk crossings could cause star formation .
we have therefore explored the process in more detail .
we prescribe the galactic potential , and allow the gc to move within it .
we also model the relative movement of a cloud of disk material on a circular orbit , as it encounters the gc crossing the disk , and show that focussing of disk material will occur for several of the gcs . the compression is on a scale of @xmath0 10pc , as suggested by wallin et al .
the extent of compression increases with gc mass and time spent by the gc in the cloud .
even if no gas is present at the time of interaction , the concentrated disk material can subsequently attract gas , leading possibly to star formation .
should gas be immediately available in the impact region , a potentially more effective mechanism exists , proposed by levy ( 2000 ) , based on shock wave compression by the gc transiting through the disk .
the star forming region in this case is estimated to be of kpc size .
we have then examined whether there is any observational evidence for star formation as a result of a gc disk crossing .
the uncertainties in the parameters in our subset of fifty - four gcs , combined with the uncertainty in the potential model are such that the impact regions are generally too extensive , subtending large angles on the sky .
for some nearby gc crossings this is not the case , however , and for ngc 3201 , ngc 6397 , and ngc 6838 , the subtended angles were smaller , as determined by monte carlo simulations . using the cds we identified a number of candidate star clusters and associations within the impact regions .
we were also able to exclude a number of clusters and associations within the 90% contour on the grounds of age or distance .
the only prior candidate for gc disk crossing star formation , the open cluster ngc 6231 , lies well outside the 90% contour of ngc 6397 impacts , in contradiction of the suggestion by rees & cudworth in wright ( 2004 ) .
we then reversed the approach , and considered whether any of the galaxy s five most prominent star super clusters ( arches , quintuplet , ngc 3603 , westerlund 1 and westerlund 2 ) could have resulted from an impact by any of the sample of fifty - four gcs , using age arguments , and the monte carlo technique applied above . from this
we found that only 7 gcs are candidates to provide impact site progenitors for the arches and quintuplet sscs , with none for the other sscs .
as a final test , we considered all 155 gcs in the harris catalogue and determined which sscs could conceivably have resulted from a gc disk crossing .
we used the criterion that the minimum speed to travel from a prior position of the sscs to the current gc position should not exceed the galactic escape velocity .
this eliminated many of the larger sample of gcs , depending on the time elapsed between disk crossing and the onset of star formation .
if the time between the disk crossing and the onset of star formation is limited to a few myr , it is possible to exclude any of the galactic gc disk crossings as a potential cause of ngc 3603 and westerlund 2 formation .
this research used the smithsonian / nasa ads , nasa s heasarc coordinate converter , and the cds at strasbourg .
dvp acknowledges the support of an stfc grant .
mike fellhauer , ignacio ferreras , chris flynn , esko gardner , jason kalirai , barbara pichardo , and kinwah wu helped us greatly by sharing their knowledge of galaxy modelling .
we are also grateful to the referee who helped us clarify some of the aspects of this work , thus improving the paper . | the crossing of the galactic disk by a globular cluster could produce star formation due to gravitational focussing or compression of disk material .
we report on simulations of the effect on disk material which reveal that the crossing can sometimes cause local gravitational focussing of disk material .
we also present the salient points of a little - known paper by levy ( 2000 ) , that shows that strong compression can result from the shock wave generated by gc disk crossing .
the main thrust of our paper is a search for remnants of disk crossings by globular clusters . using
the gravitational potential of the galaxy to locate the position of the most recent crossings of a subset of fifty - four globular clusters reveals that systematic errors and uncertainties in initial conditions limit the scope for unequivocal identification . from the subset of fifty - four ,
six possible search sites with the best constraints are retained for further scrutiny .
three of the six potentially promising search areas in the disk are from globular clusters ngc 3201 , 6397 and ngc 6838 , for which we can not rule out some observed star associations observed nearby as being remnants .
the three other of the six areas are too large to provide meaningful identification of remnants .
also , a possible remnant ( open cluster ngc6231 ) is shown not to be due to globular cluster impact , contrary to a previous report . in a more wide - ranging screening of one hundred and fifty - five globular clusters we identify which globular clusters are compatible with being responsible for the formation of any of the galaxy s five most prominent star super clusters . [ firstpage ]
galaxy : disk globular clusters : general . |
electromagnetic field enhancement in nanoplasmonic structures plays an important role in a number of applications such as nano antennas @xcite , single - photon emitters @xcite , and surface - enhanced spectroscopy techniques in general @xcite , including surface - enhanced raman spectroscopy ( sers ) @xcite .
field localization is the key to understanding and further enhancing the resolution of sers , and the use of highly engineered surface plasmonic structures has been envisioned @xcite .
strong field enhancement has so far been inherently connected to singularities @xcite .
it is commonplace to study the plasmonic response using maxwell s equations with the bulk dielectric function of the metal , with an abrupt change at the air - metal interface @xcite .
though clearly neglecting surface and confinement effects for the electrons in the metal , this is a celebrated approximation which predicts the existence of surface plasmons forming the heart of nanoplasmonics .
however , when samples involve true nano - scale feature - sizes , the validity of this approximation is challenged and new phenomena , such as non - local response and spatial dispersion @xcite and inhomogeneous electron densities due to quantum confinement @xcite are expected to play an important role . in nanoscale metallic structures
the electrons are subject to quantum confinement , which leads to a number of spectacular effects , such as the quantized conductance of atomic quantum - point contacts @xcite , friedel oscillations and confinement of electrons to quantum corrals on noble metal surfaces @xcite , and a size - dependent structure of nanocrystals @xcite .
light emission from stm has been used to reveal plasmonic properties of electrons confined to triangular islands on surfaces of noble metals @xcite .
all these studies clearly show that , at least at low temperatures , quantum confinement plays an important role leading to inhomogeneous electron densities at interfaces that differ from the piecewise constant bulk values .
the qualitative optical consequences of an inhomogeneous electron density can be explored within lindhard s _ `` local density approximation '' _ for the dielectric function @xcite @xmath0 where @xmath1 accounts for the density variations relative to the bulk density of electrons @xmath2 , with @xmath3 being the bulk plasma frequency . to mimic the ohmic damping we have included a damping rate @xmath4 and in the optical regime @xmath5 @xcite .
while several studies have considered the consequences of local density variations for the surface - plasmon dispersion relation , the overall effects have been reported to be minor and for planar surfaces the surface - plasmon resonance at @xmath6 appears unaffected by the detailed density profile at the metal - air surface @xcite .
however , it remains an open question to which extent the detailed field distribution will be affected by a smoothly varying electron density in contrast to the discontinuous drop in the electron density assumed by the bulk approximation .
zero - index phenomena in metamaterials are receiving increasing attention @xcite , and the recent results by litchinitser _
et al . _
@xcite show that for a metamaterial with the effective constitutive coefficients varying linearly , @xmath7 and @xmath8 , a pronounced field enhancement will occur at @xmath9 where @xmath10 . in this paper
we show that plasmonic structures employing nature s own metals may support large field enhancement at the zero-@xmath11 surface appearing in the vicinity of the air - metal interface , where the real part of the polarization of the dilute electron gas will exactly balance the polarization of the vacuum background .
as shown below , the particular position of the zero-@xmath11 surface depends on the frequency @xmath12 of the incident radiation , and may be either in air or in the metal , depending on the local geometry .
the basic mechanism behind the field enhancement can be understood from the continuity condition for the electrical displacement field , which stipulates that @xmath13 be continuous ( here @xmath14 is the normal vector of a surface element ) . at the zero-@xmath11 surface the normal component @xmath15 thus has to diverge to keep the product @xmath16 finite . from eq .
( [ eq : lindhard ] ) it follows that in the regions where @xmath17 vanishes , there will still be a small , but finite imaginary part : @xmath18 . for an electrical field
@xmath19 incident at an angle @xmath20 we thus expect that the normal component @xmath21 is enhanced by a factor proportional to @xmath22 on the zero-@xmath11 surface .
the intensity enhancement can then be estimated as @xmath23 we emphasize that this mechanism of field - enhancement is generic and robust as it only depends on the assumed continuous negative - to - positive variation of the dielectric function at the metal - air interface , in combination with a maxwell boundary condition @xcite , rather than on the details of the excitation of strongly localized surface - plasmon resonances with pronounced sample - to - sample fluctuations .
the low - temperature density profile for jellium is a function of @xmath24 only , with the radius of the free - electron sphere @xmath25 and @xmath26 being the bohr radius . the position - dependent electron density may be obtained self - consistently with the aid of density - functional theory @xcite , thus taking into account both exchange and correlation effects in the inhomogeneous electron gas . in ref .
@xcite , results are tabulated for a range of values of @xmath24 . the resulting density @xmath27 is shown in fig .
[ fig1](a ) for @xmath28 , a typical value for gold or silver . as seen
, @xmath29 varies continuously on the length scale of the fermi wavelength rather than exhibiting a discontinuous drop at the surface of the metal ( @xmath9 ) . in the metal ( @xmath30 )
, there are friedel oscillations near the surface , while for @xmath31 the decay of the density is caused by the finite value of the work function @xmath32 , which allows electron wave functions to penetrate into the classically forbidden region . according to eq .
( [ eq : lindhard ] ) , the dielectric function will inherit the friedel oscillations in the metal , as well as approach the value in vacuum outside the metal surface . as a consequence ,
the dielectric function passes through zero just outside the metal surface [ green line in fig .
[ fig1](a ) ] .
similar spatial variations of the dielectric function have been reported in more elaborate studies of e.g. semiconductor quantum dots @xcite and recently also in doped semiconductors @xcite . in the case of @xmath33
the zero-@xmath11 surface appears at a distance of 0.25 nm from the surface ( @xmath9 ) .
[ fig1](b ) illustrates the field enhancement in the case of a plane wave incident at an angle of @xmath34 .
the numerical results were obtained by solving the wave equation @xmath35 with the aid of a finite - element method employing an adaptive mesh algorithm ( comsol multiphysics ) . a comparison with @xmath36 in fig
. [ fig1 ] ( a ) shows that the field enhancement indeed occurs at the point where @xmath37 .
we emphasize that this phenomenon has no counterpart within the classical bulk approximation of the dielectric function . for a realistic value of the damping ( @xmath38 ) the intensity
may be enhanced by more than 3 orders of magnitude , which suggests that in a sers experiment the raman signal could be enhanced by 6 orders of magnitude @xcite without the need for any geometry - induced localized resonances . to further test the predictions of eq .
( [ scaling ] ) we have explored the dependence on the angle of incidence as well as the particular value of the plasmon damping .
[ fig2 ] ( a ) illustrates how the peak height depends on damping .
the enhancement is limited by the plasmonic damping with the slope of the dashed lines being in full accordance with the power ( -2 ) of @xmath4 in eq .
( [ scaling ] ) .
panel ( b ) illustrates the angle dependence with the dashed line showing the @xmath39 dependence of eq .
( [ scaling ] ) , indicating a pronounced enhancement for oblique incidence .
we next explore the field enhancement in the presence of strong confinement of electrons in metallic nanostructures . here
we limit ourselves to a model of non - interacting electrons in wire geometries with nanoscale cross sections @xmath40 so that the problem effectively is two - dimensional .
the energy states of the electron system are then given by @xmath41 with the transverse energy components being quantized and governed by a two - dimensional schrdinger equation @xmath42\psi_\nu(x , y)={\mathscr e}_\nu\psi_\nu(x , y).\ ] ] for the confinement potential @xmath43 we adjust the height in accordance with the work function of the metal @xmath44 , where @xmath45 is the chemical potential .
in equilibrium the states are populated according to the fermi
dirac distribution function and we find @xmath46 where @xmath47 is area and @xmath48 $ ] with @xmath49 being jonquire s polylogarithm function .
equilateral triangular nano islands have been the subject of numerous investigations @xcite and in fig .
[ fig3 ] we consider field enhancement in such a structure .
[ fig3](a ) shows results within the bulk approximation , where an artificial rounding of the corners has been added to circumvent the effect of an otherwise divergent electrical field @xcite .
clearly , field enhancement is modest since no pronounced surface plasmon resonances are excited .
[ fig3](b ) on the other hand is based on the spatially continuously varying dielectric function of eqs .
( [ eq : lindhard ] ) and ( [ eq : jonquiere ] ) and shows a pronounced field enhancement along the zero-@xmath11 surface .
we also note that while the geometry has arbitrarily sharp corners , the electron density and dielectric function themselves are still smooth , showing that quantum confinement provides a built - in smoothing even in the presence of sharp geometrical features .
[ fig4](a ) and ( b ) show in more detail the field enhancement along the lines indicated in fig .
[ fig3](b ) . for the results in fig .
[ fig4](b ) , the peak height clearly matches the one - dimensional results in panel ( b ) of fig .
[ fig2 ] for an incident angle of @xmath50 .
similarly , the result within the bulk approximation resembles typical results , see e.g. ref .
@xcite , i.e. the field enhancement is more modest . in fig .
[ fig4](a ) , the incident wave propagates almost parallel to the zero-@xmath11 surface ( @xmath51 ) , thus causing a peak that is even higher at the two right - most corners of the triangle . on the other hand , at the left - most corner
the wave has normal impact on the zero-@xmath11 surface ( @xmath52 ) so that no pronounced field enhancement is observed .
it is interesting to note that the enhancement may occur either outside [ as in fig .
[ fig1](b ) ] or inside [ as in fig . [
fig3](b ) ] the metal , depending on the local geometry and the frequency relative to the plasma frequency .
we predict that a large non - resonant field enhancement may occur at zero - epsilon surfaces near the air - metal interface .
theoretically , the enhancement occurs in a wide class of models where the discontinuous jump of the dielectric function of the usual bulk description of at an air - metal interface is replaced by a continuous function @xmath53 that has the bulk values as limiting values away from the interface .
we discussed two microscopic models for such a continuous variation at the interface .
the predicted field enhancement has therefore no counterpart in the classical bulk treatment of plasmonic field enhancement , though it is intimately related to the familiar boundary condition for the electrical displacement field .
in particular , for light impinging on an interface we discussed the dependence of the field enhancement on the angle of incoming light and on the damping in the metal . furthermore , the enhancement constitutes another mechanism for the enhancement of raman signals at metal interfaces ( sers ) .
interestingly , feigenbaum _ et al .
_ have recently reported their experimental observation of zero - index driven field enhancement at the surface of conducting oxides ( i.e. with lower electron density compared to good metals ) @xcite .
we emphasize that the decay of the electron density into vacuum ( or more generally into a dielectric ) could have important implications also for nano - gap structures ( such as the gap formed between the corners of two nearby triangles @xcite ) , where quantum tunneling of electrons will introduce a small but finite electron density inside the gap , thus also modifying the local dielectric function and the ability to support short - circuiting currents @xcite .
we thank peter nordlander , natalia m. litchinitser , and vladimir m. shalaev for useful discussions .
this work is financially supported by the danish council for strategic research through the strategic program for young researcher ( grant no : 2117 - 05 - 0037 ) , the danish research council for technology and production sciences ( grants no : 274 - 07 - 0080 & 274 - 07 - 0379 ) , as well as the fidipro program of the finnish academy .
o. perez - gonzalez , n. zabala , a. g. borisov , n. j. halas , p. nordlander , and j. aizpurua , `` optical spectroscopy of conductive junctions in plasmonic cavities , '' _ nano lett .
_ * 10*(8 ) , 3090 3095 ( 2010 ) .
j. v. lauritsen , j. kibsgaard , s. helveg , h. topse , b. s. clausen , e. lgsgaard , and f. besenbacher , `` size - dependent structure of o@xmath54 nanocrystals , '' _ nat . nanotechnol . _ * 2*(1 ) , 53 58 ( 2007 ) . our qualitative conclusions do not change if a more sophisticated form for the dielectric function is adopted , such as the charge - conserving form due to n. d. mermin , _ phys .
b _ * 1*(5 ) , 2362 ( 1970 ) .
r. liu , q. cheng , t. hand , j. j. mock , t. j. cui , s. a. cummer , and d. r. smith , `` experimental demonstration of electromagnetic tunneling through an epsilon - near - zero metamaterial at microwave frequencies , '' _ phys .
lett . _ * 100*(2 ) , 023903 ( 2008 ) .
b. edwards , a. al , m. e. young , m. silveirinha , and n. engheta , `` experimental verification of epsilon - near - zero metamaterial coupling and energy squeezing using a microwave waveguide , '' _ phys . rev .
lett . _ * 100*(3 ) , 033903 ( 2008 ) .
r. j. pollard , a. murphy , w. r. hendren , p. r. evans , r. atkinson , g. a. wurtz , a. v. zayats , and v. a. podolskiy , `` optical nonlocalities and additional waves in epsilon - near - zero metamaterials , '' _ phys . rev
. lett . _ * 102*(12 ) , 127405 ( 2009 ) .
a. sundaramurthy , p. j. schuck , n. r. conley , d. p. fromm , g. s. kino , and w. e. moerner , `` toward nanometer - scale optical photolithography : utilizing the near - field of bowtie optical nanoantennas , '' _ nano lett .
_ * 6*(3 ) , 355360 ( 2006 ) .
j. nelayah , m. kociak , o. stphan , f. j. garca de abajo , m. tenc , l. henrard , d. taverna , i. pastoriza - santos , l. m. liz - marzn , and c. colliex , `` mapping surface plasmons on a single metallic nanoparticle , '' _ nat .
phys . _ * 3*(5 ) , 348353 ( 2007 ) . | we point out an apparently overlooked consequence of the boundary conditions obeyed by the electric displacement vector at air - metal interfaces : the continuity of the normal component combined with the quantum mechanical penetration of the electron gas in the air implies the existence of a surface on which the dielectric function vanishes .
this , in turn , leads to an enhancement of the normal component of the total electric field .
we study this effect for a planar metal surface , with the inhomogenous electron density accounted for by a jellium model .
we also illustrate the effect for equilateral triangular nanoislands via numerical solutions of the appropriate maxwell equations , and show that the field enhancement is several orders of magnitude larger than what the conventional theory predicts . |
the orbital degree of freedom of the electron has attracted considerable attention in recent years due to the rich variety of electronic , magnetic , and transport properties exhibited by orbitally degenerate systems such as the ferromagnetic manganites , which have highlighted the interplay between spin and orbital degrees of freedom in these correlated electron systems.@xcite orbital fluctuations , correlations , and orderings have been observed in raman spectroscopic studies@xcite of orbiton modes in @xmath7 , polarization - contrast - microscopy studies@xcite of @xmath8 , magnetic susceptibility and inelastic neutron scattering studies@xcite of @xmath9 , and resonant inelastic soft x - ray scattering studies@xcite of @xmath10 and @xmath11 . a new detection method for orbital structures and ordering based on spectroscopic imaging scanning tunneling microscopy is of strong current interest@xcite in orbitally active metallic systems such as strontium ruthenates and iron pnictide superconductors .
the role of orbital fluctuations on magnetic couplings and excitations is of strong current interest in view of the several zone - boundary anomalies observed in spin - wave excitation measurements in the metallic ferromagnetic phase of colossal magnetoresistive ( cmr ) manganites.@xcite the presence of short - range dynamical orbital fluctuations has been suggested in neutron scattering studies of ferromagnetic metallic manganite @xmath12.@xcite these observations are of the crucial importance for a quantitative understanding of the carrier - induced spin - spin interactions , magnon excitations , and magnon damping , and have highlighted possible limitations of existing theoretical approaches .
for example , the observed magnon dispersion in the @xmath6-x direction shows significant softening near the zone boundary , indicating non - heisenberg behaviour usually modeled by including a fourth neighbour interaction term @xmath13 , and highlighting the limitation of the double - exchange model .
similarly , the prediction of magnon - phonon coupling as the origin of magnon damping@xcite and of disorder as the origin of zone - boundary anomalous softening@xcite have been questioned in recent experiments.@xcite furthermore , the dramatic difference in the sensitivity of long - wavelength and zone - boundary magnon modes on the density of mobile charge carriers has emerged as one of the most puzzling feature . observed for a finite range of carrier concentrations , while the spin stiffness remains almost constant , the anomalous softening and broadening of the zone - boundary modes show substantial enhancement with increasing hole concentration.@xcite theoretically , the role of orbital - lattice fluctuations and correlations on magnetic couplings and excitations has been investigated within an orbitally degenerate double exchange model with an inter - orbital interaction @xmath14 and the jahn - teller coupling.@xcite based on a strong coupling expansion , this approach is restricted to the strong coupling limit @xmath15 .
the final calculations for the magnon self energy , carried out in terms of a phenomenological parameter , show significant zone - boundary magnon softening only for ferromagnetic orbital correlations , and extremely close to the orbital ordering instability .
if orbital fluctuations have signature effects on magnetic excitations in a ferromagnet with orbital degree of freedom , they can be probed indirectly through neutron scattering studies .
a detailed investigation of the orbital fluctuation magnon self energy is therefore of strong current interest , especially with respect to dependence on inter - orbital interaction strength , band filling , and different orbital fluctuation modes . in this paper
we will present a theory for spin - orbital coupling and magnon self energy , and examine how the correlated motion of electrons in the presence of strong orbital correlations near the orbital ordering instability influences magnetic couplings and excitations .
we will employ a diagrammatic approach which allows interpolation in the full range of interaction strength from weak to strong coupling . in this approach correlation effects in the form of self energy and vertex corrections
are incorporated systematically so that the goldstone mode is preserved order by order .
based on a systematic inverse - degeneracy expansion scheme,@xcite the approach has been used recently to study spin - charge coupling effects , which give rise to strong magnon energy softening , damping , and non - heisenberg behaviour in metallic ferromagnets.@xcite the present work will also extend the recent investigation into role of orbital degeneracy and hund s coupling on magnetic couplings and excitations in a band ferromagnet.@xcite orbital degeneracy and hund s coupling were shown to enhance ferromagnetism by strongly suppressing correlation - induced quantum corrections to spin stiffness and magnon energies . an effective quantum parameter was obtained for determining the magnitude of quantum corrections , and the theory was applied to calculate the spin stiffness for a realistic multi - orbital system such as iron .
we will show here that the spin stiffness remains essentially unaffected by the interaction @xmath14 due to the non - heisenberg @xmath5 behaviour of the magnon self energy resulting from orbital fluctuations and correlations . in manganites , an important role
is also played by the cooperative jahn - teller distortion of o@xmath16 ions which lifts the two - fold degeneracy of @xmath17 electronic levels of mn due to a combination of orbital geometry and electrostatic repulsion , leading to staggered orbital correlations .
this is qualitatively similar to the local orbital moment and staggered orbital correlations introduced by the inter - orbital density interaction @xmath18 which relatively pushes up the @xmath19 orbital energy if the @xmath20 orbital density @xmath21 is more than average , thus self consistently lifting the orbital degeneracy .
therefore , orbital correlations and fluctuations due to dynamical jahn - teller distortion can be qualitatively treated in terms of an effective inter - orbital interaction .
the structure of the paper is as follows .
starting with a degenerate two - orbital hubbard model including an inter - orbital interaction @xmath14 , the first order quantum correction diagrams for the irreducible particle - hole propagator are obtained in section ii . as basic ingredients in the diagrammatics , spin and orbital fluctuations
are briefly discussed in section iii .
the magnon self energy contributions due to orbital fluctuations and spin - orbital coupling are obtained in sections iv and v. the interplay between magnetic and charge contributions to the spin - orbital interaction vertex is discussed in section vi , and orbital fluctuations near @xmath0 are shown to yield strong zone - boundary magnon softening .
extension to finite hund s coupling @xmath22 and the ferromagnetic kondo lattice model are discussed in sections vii and viii , and conclusions are presented in section ix .
we will consider a two - orbital hubbard model @xmath23 on a simple cubic lattice with two orbitals ( labeled by @xmath24 ) per site and intra - orbital nearest - neighbor hopping @xmath25 .
the coulomb interaction matrix elements included here are the intra - orbital interaction @xmath26 , the inter - orbital density interaction @xmath14 , and the inter - orbital exchange interaction ( hund s coupling ) @xmath22 .
the last term represents the transverse part ( @xmath27 ) of the hund s coupling , and the density interaction term yields the longitudinal part @xmath28 , so that altogether the hund s coupling term has the form @xmath29 .
the hamiltonian therefore possesses continuous spin rotation symmetry , and hence the goldstone mode must exist in the spontaneously broken symmetry state .
hund s coupling has been shown to strongly enhance ferromagnetism in an orbitally degenerate system by strongly suppressing the correlation - induced quantum corrections.@xcite an effective quantum parameter @xmath30/[u+({\cal n}-1)j]^2 $ ] was obtained for determining the magnitude of quantum corrections for an @xmath31-orbital system , in analogy with @xmath32 for quantum spin systems .
the rapid decrease of this quantum factor from 1 to @xmath33 as @xmath34 increases from 0 to 1 results in strong suppression of quantum corrections and hence significant stabilization of ferromagnetism by hund s coupling , particularly for large @xmath31 . in order to highlight the role of inter - orbital coulomb interaction @xmath14 and orbital fluctuations on magnetic couplings and excitations in this paper
, we will first set @xmath35 .
the case of finite hund s coupling will be treated later in section vii . in a band ferromagnet ,
all information regarding carrier - induced spin interactions @xmath36 and excitations are contained in the irreducible particle - hole propagator @xmath37 , which then yields the exact transverse spin fluctuation ( magnon ) propagator:@xcite @xmath38 our approach is to incorporate correlation effects in @xmath37 in the form of self - energy and vertex corrections using a systematic expansion @xmath39 which preserves the goldstone mode order by order . rooted within an inverse - degeneracy expansion scheme ,
this systematic approach is non - perturbative with respect to the interaction terms and therefore yields a controlled approximation which remains valid in the strong coupling limit .
contributions to the first order quantum correction @xmath40 due to the hubbard interaction @xmath26 and hund s coupling @xmath22 have been discussed earlier.@xcite the additional first order diagrams for @xmath41 arising from the inter - orbital interaction @xmath14 are shown in fig .
1 . the diagrams shown here are for a saturated ferromagnet in which minority ( @xmath42 ) spin particle - hole fluctuations are absent . here
1(a ) and ( b ) represent quantum corrections due to electronic self energy renormalization by orbital fluctuations , fig .
1(c ) represents the corresponding vertex correction , and fig .
1(d ) represents vertex corrections involving coupling between transverse spin and orbital fluctuations .
the vertex correction diagrams as in fig .
1(d ) ( nine such diagrams ) can be represented in terms of an effective spin - orbital interaction vertex @xmath43 as shown in fig .
the spin - orbital interaction vertex has three contributions involving three - fermion vertices , as shown in fig .
the missing fourth diagram vanishes because of the assumption of complete polarization .
as the goldstone - mode condition @xmath44 at @xmath45 is already exhausted by the zeroth - order ( classical ) term @xmath46 , the sum of the higher order ( quantum ) terms @xmath47 must exactly vanish at @xmath45 . for this cancellation to hold for
arbitrary @xmath26 , @xmath22 , and @xmath14 , each higher order term @xmath48 in the expansion must individually vanish , implying that the goldstone mode is preserved order by order .
we will demonstrate this exact cancellation explicitly for the new contributions due to @xmath14 in the first - order quantum correction @xmath40 .
systematics in our two - orbital model can be formally introduced , in analogy with the inverse - degeneracy ( @xmath33 ) expansion for the hubbard model , by : i ) treating the two physical orbitals @xmath24 as pseudo spins , ii ) introducing @xmath31 pseudo orbitals ( @xmath49 ) for each pseudo spin , and iii ) generalizing the inter - orbital density interaction to @xmath50 .
now , each interaction line @xmath14 yields a factor @xmath33 and each bubble yields a factor @xmath31 from the summation over pseudo orbitals , resulting in an overall @xmath33 factor for the bubble series , and an overall @xmath51 factor in the @xmath52-order quantum correction @xmath48 .
the diagrammatic expansion above involves spin and orbital fluctuation propagators , the characteristic energy and momentum distribution of which are important in view of the spin - orbital coupling investigated in this work . the ladder series in fig .
1(d ) yields the effective intra - orbital transverse spin interaction : @xmath53 in terms of the rpa - level magnon propagator @xmath54 , having an advanced pole in the saturated ferromagnetic state ( @xmath55 ) . here
@xmath56 is the bare antiparallel - spin particle - hole propagator , @xmath57 and @xmath58 are the magnon - mode amplitude and energy , and the small weight of gapped stoner excitations has been neglected for simplicity . similarly , the bubble series in fig .
1 involving odd number of bubbles yields , in terms of the rpa - level orbital fluctuation propagator , the effective intra - orbital density interaction : @xmath59 near the orbital ordering instability where @xmath60 . here
@xmath61 is the bare spin-@xmath62 particle - hole propagator .
the orbital fluctuation propagator is symmetric @xmath63 , with a spectral representation : @xmath64}{\omega - \omega ' + i \eta}\ ] ] for its retarded part , with a continuum distribution over the orbital fluctuation energy @xmath65 .
exactly at quarter filling @xmath66 per orbital ) , the orbital fluctuation propagator diverges at @xmath67 in the absence of any nnn hopping @xmath68 terms which destroy fermi surface nesting , indicating instability towards staggered orbital ordering . with increasing hole doping , the spectral function peak shifts below @xmath69 .
3(a ) shows the momentum integrated orbital fluctuation ( para - orbiton ) spectral function @xmath70 with increasing interaction strength @xmath14 . here , and in the following
, we have set the hopping term @xmath71 as the unit of energy , where @xmath72 is the bandwidth . in analogy with the well - known para - magnon response with approaching magnetic instability ,
the para - orbiton energy scale is strongly suppressed from order bandwidth in the weak - coupling regime to relatively very low energies near the orbital ordering instability .
3(b ) shows the momentum dependence of the low - energy @xmath73 integrated part of the orbital fluctuation spectral function near the r point @xmath69 , which shows that low - energy orbital fluctuations are concentrated near the wavevector @xmath69 corresponding to staggered orbital fluctuations . with increasing doping away from quarter filling , the peak shifts below @xmath69 , indicating incommensurate fluctuations .
we will first consider diagrams for @xmath41 in fig .
1 ( a , b , c ) involving electronic self energy corrections due to orbital fluctuations and the corresponding vertex correction .
absent in the single - orbital case , these diagrams are characteristic of orbital degeneracy , inter - orbital interaction , and orbital fluctuations , and strongly influence magnetic couplings and excitations through electronic band renormalization , particularly in vicinity of the orbital ordering instability . the vertex correction diagrams ( d ) involving both spin and orbital fluctuations will be discussed in the next section .
quantum corrections to the irreducible particle - hole propagator @xmath41 in eq .
( 2 ) yield the magnon self energy : @xmath74 \ ; , \ ] ] in terms of which the magnon propagator @xmath75 $ ] . the first - order magnon self energy corresponding to diagrams in fig .
1(a , b , c ) involving only orbital fluctuations is then obtained by summing over the bosonic degrees of freedom of the orbital fluctuations : @xmath76 = m v^2 \sum_{\bf q } \int_{-\infty } ^{\infty } \frac{d\omega}{2\pi i } \ ; \chi_{\rm orb } ( { \bf q - q},\omega-\omega ) \ ; \gamma_4 ( { \bf q , q},\omega,\omega)\ ] ] where @xmath77 is the four - fermion vertex obtained by integrating out the fermionic degrees of freedom in the diagrams for @xmath41 shown in fig .
1(a , b , c ) .
as will be further discussed in the following three subsections , this orbital fluctuation magnon self energy physically represents contributions due to ( i ) coupling between stoner excitations and orbital fluctuations , and ( ii ) self - energy corrections involving band - energy renormalization and spectral - weight transfer . using the spectral representation for the orbital fluctuation propagator ( retarded part , since @xmath77 has only advanced poles with respect to @xmath78 ) ,
we obtain : @xmath79 where the average four - fermion vertex @xmath80 denotes averaging over orbital fluctuation modes @xmath81 .
evaluation of the four - fermion vertex @xmath77 , resolved into different contributions corresponding to distinct physical mechanisms , is discussed below .
term by term , the four - fermion vertex @xmath77 vanishes identically for @xmath45 , in accord with the goldstone mode . in the single - orbital case ,
the magnon self energy due to spin - charge coupling included a stoner contribution representing coupling of charge excitations with the gapped part of spin excitations.@xcite in analogy , diagrams fig .
1(a , b , c ) yield contributions which represent a stoner - orbital coupling : @xmath82 in which the first term represents the stoner excitation mode @xmath83 and the quadratic term is the stoner - orbital interaction vertex , which involves only magnetic energy denominators . here
@xmath84 are the ferromagnetic state band energies for the two spins in terms of the free - particle energy @xmath85 for the sc lattice and the exchange splitting @xmath86 .
the band energy superscripts @xmath87 refer to particle ( hole ) states .
there is no restriction on the energy @xmath88 in eq .
( 9 ) as both particle and hole states contribute .
of the four terms in this quadratic interaction vertex , the two square terms arise from diagrams fig .
1(a ) and ( b ) , while the cross terms are from diagram fig .
1(c ) ; the characteristic quadratic structure therefore stems from orbital fluctuations renormalizing electrons of both spins , and is clearly absent in the single - orbital case involving spin - down renormalization only .
this intrinsic quadratic structure resulting from orbital degeneracy yields a characteristic non - heisenberg @xmath5 magnon self energy , resulting in no spin stiffness reduction but strong zone - boundary magnon energy reduction .
due to exchange of inter - orbital fluctuations in the diagram fig .
1(a ) involving intermediate spin-@xmath62 states , the spin-@xmath62 hole ( particle ) energies are pulled down ( pushed up ) , increasing the particle - hole energy gap , and thereby suppressing the particle - hole propagator @xmath41 . including the corresponding contributions from the vertex correction diagram fig .
1(c ) , we obtain the electronic band renormalization contribution : @xmath89 involving one charge and two magnetic energy denominators .
the finite infinitesimal term @xmath90 as in eq .
( 9 ) has been dropped for compactness . the electronic self energy correction in diagram fig .
1(a ) also results in spectral - weight transfer and redistribution between occupied and unoccupied spin-@xmath62 states . however , there is no net change in occupancy and magnetization .
the corresponding spectral weight transfer contribution : @xmath91 involves one magnetic and two charge energy denominators .
the first ( negative ) contribution corresponds to loss of spin-@xmath62 hole spectral weight due to transfer to particle states , and the second ( positive ) contribution corresponds to the reverse process .
singular contributions in eqs .
( 9 - 11 ) for @xmath77 exactly cancel out .
for example , the most singular contribution in eq .
( 9 ) involving two powers of the vanishing energy denominator ( @xmath92 ) exactly cancels with the corresponding contribution from eq . (
similarly , the next most singular contributions in eqs .
( 9 - 11 ) also exactly cancel out .
the average @xmath80 of the four - fermion vertex over orbital fluctuation modes directly yields the magnon self energy from eq .
( 8) . since orbital fluctuations peak below @xmath69 for finite doping [ fig .
3 ] , the vertex @xmath93 was estimated by averaging over a selected @xmath94 region ( @xmath95 ) near @xmath69 , with @xmath96 assumed flat inside and zero outside . the @xmath97 dependence of the averaged four - fermion vertex @xmath98 in the @xmath6-x direction of the brillouin zone is shown in fig .
the band contribution is negative due to renormalization of band energies by orbital fluctuations , as discussed above .
the band contribution rapidly diminishes in the strong coupling limit [ fig .
4(b ) ] , as does the stoner contribution .
the spectral contribution survives in the strong coupling limit .
all contributions have strongly non - heisenberg character , with negligible magnitude at small @xmath97 but rising sharply near the zone boundary , implying no spin stiffness correction but appreciable zone boundary magnon softening .
orbital fluctuation modes near @xmath99 etc . also yield significant non - heisenberg character to the interaction vertex @xmath100 , and the three contributions exhibit similar behaviour .
how does the orbital fluctuation magnon self energy compare with the bare magnon energy ? taking the orbital fluctuation energy @xmath65 to be negligible in comparison to the bandwidth near the orbital ordering instability , and the estimated average @xmath101 near the zone boundary from fig .
4(c ) , we obtain ( for @xmath102 and @xmath103 ) : @xmath104 which is comparable to the bare zone - boundary magnon energy @xmath105 for realistic strength of the inter - orbital interaction .
in the previous section , we considered the diagrams of fig
. 1 ( a , b , c ) involving electronic self - energy corrections due to orbital fluctuations . we will now examine the vertex correction diagrams of fig .
1(d ) representing spin - orbital coupling , which are particularly important near the orbital ordering instability where orbital fluctuations are soft .
in contrast to the single - orbital case where self energy and vertex correction diagrams were of qualitatively similar order,@xcite introduction of the orbital degree of freedom lifts this constraint and allows qualitatively independent self energy and vertex correction contributions .
the corresponding first - order magnon self energy : @xmath106 [ \gamma_3 ( { \bf q , q},\omega,\omega)]^2 [ v_{\rm eff } ^{\alpha\alpha } ( { \bf q - q},\omega-\omega ) ] \nonumber \\ & = & m \sum_{\bf q}\int\frac{d\omega}{2\pi i } [ \chi_{\rm sp } ( { \bf q},\omega ) ] [ \gamma_{\rm sp - orb } ( { \bf q , q},\omega,\omega)]^2 [ \chi_{\rm orb } ( { \bf q - q},\omega-\omega ) ] \end{aligned}\ ] ] where @xmath54 and @xmath107 are the spin and orbital fluctuation propagators ( section iii ) , and @xmath108 represents the spin - orbital interaction vertex in terms of the three - fermion vertex @xmath109 , evaluation of which is discussed in the appendix . in analogy with the spin - charge coupling process,@xcite
this correlation - induced spin - orbital coupling is analogous to a second - order raman scattering process in which the magnon @xmath110 scatters into an intermediate - state magnon @xmath83 along with an internal orbital excitation @xmath111 , leading to significant magnon energy renormalization and magnon damping .
the @xmath112 integrals in the above equation represent integrating out the bosonic ( both spin and orbital ) degrees of freedom . as the magnon propagator @xmath113 $ ] is purely advanced in nature
, only the retarded part of the product @xmath114 ^ 2 [ \chi_{\rm orb}]$ ] contributes in the @xmath78 integral . for simplicity ,
considering the dominant contribution to the spectral representation of this product coming from the orbital fluctuation propagator , from eqs .
( 13 ) , ( 3 ) , ( 5 ) , and the symmetry property given above eq .
( 5 ) , we obtain : @xmath115 ^ 2 \int_0^\infty \frac{d\omega'}{\pi } \ ;
\frac{{\rm i m } \chi_{\rm orb}({\bf q - q},\omega')}{\omega - \omega - \omega ' + i \eta } \nonumber \\ & = & m^2 \sum_{\bf q } \int_0^\infty \frac{d\omega'}{\pi } \ ; \frac { [ \gamma_{\rm sp - orb } ( { \bf q , q},\omega,\omega)]^2 } { \omega_{\bf q}^0 + \omega + \omega ' - i \eta } { \rm im}\chi_{\rm orb}({\bf q - q},\omega')\end{aligned}\ ] ] due to magnon decay into internal spin and orbital excitations ( @xmath116 ) , the above magnon self energy yields a finite imaginary part representing finite magnon damping and linewidth , as discussed earlier for spin - charge coupling.@xcite an approximate evaluation of the resulting spin - orbital magnon self energy illustrates the importance of the orbital fluctuation softening near the orbital - ordering instability . with @xmath117 and @xmath118
representing characteristic orbital and spin fluctuation energy scales , we obtain ( for @xmath119 ) : @xmath120 ^ 2 \rangle_{{\bf q'},\omega ' } } { \omega_{\rm spin } + \omega_{\rm orb } - i \eta } \ ] ] where the angular brackets @xmath121 again refer to averaging over the orbital fluctuation modes @xmath122 , as in eq .
far from the orbital ordering instability , the orbital fluctuation energy @xmath123 is of order bandwidth @xmath72 , which strongly suppresses the magnon self energy .
however , near the orbital - ordering instability , spin - orbital coupling becomes important due to the relatively much smaller energy denominator @xmath124 . the spin - orbital interaction vertex @xmath43 is obtained by integrating out the fermion degrees of freedom in the three - fermion interaction vertices .
this interaction vertex explicitly vanishes at momentum @xmath45 in accordance with the goldstone mode requirement , and yields the dominant @xmath97 dependence of the magnon self energy . in order to illustrate the characteristic non - heisenberg character of the interaction vertex
, we consider its magnetic part with energy denominators involving the stoner gap .
this term qualitatively differs from the charge part of the vertex with energy denominators involving excitation energies of order bandwidth .
evaluation of the three - fermion vertices contributing to @xmath43 is discussed in the appendix . for the purely magnetic part
, we obtain : @xmath125\ ] ] where there is no restriction on the @xmath126 energies .
the leading order @xmath97 dependence of the above term is approximately @xmath127 as shown below . for a fixed orbital fluctuation momentum @xmath128 , with @xmath94 close to @xmath69 corresponding to staggered orbital fluctuations , expansion in powers of @xmath129 of the square - bracket terms in eq .
( 16 ) yields a contribution @xmath130 to leading order .
the formally second - order structure of the spin - orbital coupling magnon self energy ( eq .
15 ) involving @xmath114 ^ 2 $ ] therefore directly yields an intrinsically non - heisenberg @xmath5 contribution to the magnon self energy , which yields no spin stiffness quantum correction , but significant magnon energy reduction and damping near the zone boundary , and therefore accounts for the observed zone - boundary magnon anomalies .
the @xmath97 dependence of the magnetic part @xmath131 ^ 2 $ ] of the spin - orbital interaction term is shown in fig .
being small in comparison to the stoner gap , the boson energies @xmath132 and @xmath78 in eq . (
16 ) were set to zero , and the momentum @xmath94 was selected in a range near @xmath69 , as in fig .
while @xmath131 ^ 2 $ ] shows strong anomalous momentum dependence for orbital fluctuation modes near @xmath69 , when both magnetic and charge terms ( see appendix ) are included , the net contribution from these modes is small due to a cancellation , as shown in fig .
in contrast , for orbital fluctuation modes with momentum @xmath94 near @xmath0 , the above cancellation is avoided as the magnetic contribution to @xmath43 is small .
we find that the total spin - orbital interaction term @xmath114 ^ 2 $ ] exhibits a strong anomalous momentum dependence [ fig .
5(b ) ] , and that it is strongly enhanced with increasing hole doping , as the band filling changes from 0.5 ( quarter filling ) to 0.25 ( one - eighth filling ) . in order
to quantitatively examine the effect of this anomalous momentum behaviour of @xmath133 on the magnon dispersion , the magnon self energy was evaluated approximately using eq .
( 15 ) . on averaging @xmath114 ^ 2 $ ] over @xmath94 ,
we find that the anomalous momentum behaviour remains qualitatively similar for @xmath134 and drops sharply for @xmath135 , whereas in the @xmath136 plane the dominant and qualitatively similar contribution comes from diagonal strips along @xmath137 , yielding a factor of @xmath138 on planar averaging , resulting in an overall phase - space factor of @xmath139 .
the magnon self energy was therefore estimated using eq .
( 15 ) with @xmath133 as obtained in fig .
5(b ) for the orbital fluctuation mode @xmath0 , and including the above phase - space factor ( taken as 1/10 in our calculations ) to account for momentum averaging over orbital fluctuation modes .
also , the bare magnon energy @xmath140 and the orbital fluctuation energy @xmath123 were taken to be of order @xmath25 .
the renormalized magnon energy : @xmath141 is shown in fig . 6 for different band fillings @xmath142 .
it is evident that while the spin stiffness remains essentially unchanged , the magnon self energy at the zone - boundary becomes comparable to the bare magnon energy for realistic values of the inter - orbital interaction @xmath14 , resulting in a zone - boundary softening which is strongly enhanced with hole doping . in the presence of staggered orbital correlations such as near @xmath0 , weighted averaging over @xmath94 with a peaked orbital spectral function will further enhance @xmath143 ^ 2 \rangle$ ] and therefore the anomalous magnon self energy .
the orbital fluctuation modes near @xmath0 correspond to period @xmath1 planar staggered correlations , as shown in fig
such orbital correlations have been observed in the ce - type charge - ordered phase of the half - doped @xmath144 manganites such as the narrow - band compounds like @xmath145 and @xmath146,@xcite and the layered material @xmath147 in which magnetic excitations were found to be dominated by ferromagnetic couplings.@xcite therefore , as @xmath148 approaches 0.5 , orbital fluctuation modes near @xmath0 may form the dominant contribution to the low - energy part of the orbital fluctuation spectral function .
the different layers in @xmath147 are magnetically decoupled due to negligible interlayer couplings .
extension of the present investigation of spin - orbital coupling to the two - dimensional case is therefore of interest with respect to renormalization of magnetic couplings . in the case of spin - charge coupling ,
the renormalized magnon dispersion for a square lattice does show strong zone - boundary softening near @xmath149 and @xmath150 , while the magnon energy near @xmath151 remains undiminished , indicating softening of the nearest - neighbour ferromagnetic bonds but strong ferromagnetic correlations along the diagonal directions.@xcite orbital fluctuation modes near @xmath152 and @xmath153 were also found to yield significant contribution to the spin - orbital interaction vertex .
involving similar density @xmath154 on all sites , such configurations should , however , be relatively suppressed by an intersite density interaction @xmath155 .
in contrast , allowing for reduced density on alternating `` empty '' sites with vanishing orbital `` magnetization '' , configurations corresponding to modes near @xmath0 with period @xmath1 orbital correlations minimize the intersite interaction energy @xmath156 , and would therefore be relatively more important .
so far we had set the hund s coupling @xmath35 in order to highlight the role of inter - orbital interaction and fluctuation . for finite @xmath22 , it is convenient to proceed in two steps .
the part @xmath157 of the inter - orbital interaction @xmath14 together with hund s coupling effectively amounts to a purely magnetic interaction @xmath158 , and has been investigated earlier.@xcite this is because for @xmath157 , the inter - orbital interaction @xmath159 acts only between opposite - spin electrons , and so the resulting diagrammatics is similar to the hubbard interaction case .
the remaining part @xmath160 is purely non - magnetic , and yields diagrammatic contributions essentially as in section ii .
thus , corresponding to fig .
1 ( a , b , c ) diagrams involving the bubble series for the orbital fluctuation propagator , the magnon self energy is obtained by simply replacing @xmath14 by @xmath160 in eq .
( 7 ) . for the spin - orbital coupling magnon energy , however , the interaction ladders in diagrams fig . 1(d ) and fig . 2 now involve hund s coupling @xmath22 as well , and since the transverse part of @xmath22 flips the orbital index in the ladder series , there are now two contributions to the irreducible particle - hole propagator : @xmath161 ^ 2 v_{\rm eff } ^{\alpha\mu } ( { \bf q - q},\omega-\omega)\ ] ] involving intra @xmath162 and inter @xmath163 orbital spin and orbital fluctuations . involving ladders of @xmath26 and @xmath22 , the effective intra and inter - orbital transverse spin interactions : @xmath164 \approx \frac{(u+j)^2}{2 } \left [ \frac{\chi_0}{1-u^+ \chi_0 } \pm \frac{\chi_0}{1-u^- \chi_0 } \right ] \ ] ] can be expressed as in - phase @xmath162 and out - of - phase @xmath163 combinations of the acoustic and optical branches,@xcite with @xmath165 .
similarly , the effective intra and inter - orbital density interactions : @xmath166 involve odd and even number of bubbles .
now , in the investigation of role of hund s coupling on quantum corrections,@xcite it was shown that the inter - orbital component @xmath167 yields a small @xmath78-integrated contribution due to partial cancellation from the out - of - phase combination of the acoustic and optical modes , essentially reflecting an inter - orbital incoherence .
therefore , the inter - orbital component @xmath168 should be relatively small , and the intra - orbital component @xmath169 is approximately given by eq .
( 15 ) , with @xmath14 replaced by @xmath160 and @xmath26 replaced by @xmath170 . due to the purely opposite - spin density interaction @xmath159 , the interaction line @xmath26 connected to the bubble in the third diagram in fig .
2(b ) for the three - fermion vertex @xmath109 is also replaced by @xmath170 .
this is consistent with the enhanced exchange splitting to @xmath171 in the @xmath172 energy denominator , which changes the magnon pole condition to @xmath173 , and ensures that the three - fermion vertex @xmath109 exactly vanishes at @xmath45 in accordance with the goldstone - mode condition ( see appendix ) .
magnetic couplings and excitations in ferromagnetic manganites have been theoretically investigated using the ferromagnetic kondo lattice model ( fklm ) and its strong coupling limit , the double exchange ferromagnet . in this model ,
the @xmath174 core spins due to localized @xmath175 electrons of the magnetic mn@xmath176 ions are exchange coupled to the mobile @xmath17 band electrons , represented by an interaction term @xmath177 , with @xmath178 of order bandwidth @xmath72 .
the fermionic representation approach for evaluating magnon self energy corrections in the fklm , which allows conventional diagrammatic tools to be employed for evaluating quantum corrections beyond the leading order,@xcite can be readily extended to include effects of orbital fluctuations . with
an inter - orbital interaction @xmath14 included between the two degenerate @xmath17 orbitals , the fklm magnon self energy due to spin - orbital coupling can be directly obtained from our magnon renormalization analysis of fig .
the required correspondence is : @xmath179 and @xmath180 , so that the exchange splitting @xmath181 .
the fklm magnon self energy is thus obtained from our calculated hubbard model result using a multiplicative factor @xmath182 . with @xmath183 , @xmath102 , @xmath184 , and @xmath185
, we obtain @xmath186 .
as the fklm bare magnon energy @xmath187 is smaller than the hubbard model bare magnon energy @xmath188 by roughly the same factor , the fklm renormalized magnon energy @xmath189 will also be as in fig .
6 , only scaled down by a factor ( 1/6 ) . taking the hopping energy scale @xmath190mev corresponding to a realistic bandwidth @xmath191ev , the magnon energy scale in fig .
6 is @xmath192mev for ferromagnetic manganites , in agreement with the measured magnon energies.@xcite since the fklm magnon self energy goes as the fourth power of @xmath193 explicitly , the anomalous zone - boundary softening effect should be especially pronounced in narrow - band systems .
indeed , zone - boundary magnon softening is clearly seen to occur in the relatively low-@xmath194 or narrow - band materials , and broad - band materials such as @xmath195 show nearly heisenberg behavior , in agreement with this prediction .
systematic studies of doping dependence of the zone - boundary magnon softening indicates that the higher the doping level , the larger the zone - boundary softening.@xcite furthermore , doping dependence of spin dynamics indicates that the measure spin stiffness @xmath196 mev @xmath197 remains essentially unchanged , while the zone - boundary magnon softening ( denoted by the ratio @xmath198 ) is found to be enhanced linearly with increasing doping .
the correlated motion of electrons in the presence of strong orbital fluctuations and correlations was investigated in an orbitally degenerate ferromagnet with two orbitals per site with respect to magnetic couplings and excitations .
a systematic goldstone - mode - preserving approach was employed to incorporate correlation effects in the form of self energy and vertex corrections , so that both long - wavelength and zone - boundary magnon modes could be studied on an equal footing .
our investigation focussed on the anomalous momentum dependence of the three- and four - fermion interaction vertices which determine the magnon self energy and the role of different orbital fluctuation modes , particularly near the orbital ordering instability where orbital fluctuations are relatively soft .
orbital fluctuations were generically shown to impart an intrinsically non - heisenberg @xmath199 character to the magnon self energy in the @xmath6-x direction of interest .
this generic behaviour was ascribed to a quadratic structure of the spin - orbital interaction vertex resulting from a new class of diagrammatic contributions associated with the orbital degree of freedom .
these diagrams are absent in the single - orbital case , the essential difference being that orbital fluctuations couple to electrons of both spin .
the absence of any @xmath200 contribution in this non - heisenberg magnon self energy for small @xmath97 implies that the spin stiffness is not renormalized by orbital fluctuations generated by the inter - orbital density interaction @xmath14 . in a multi - orbital ferromagnet , with intra - orbital coulomb interaction @xmath26 , inter - orbital interaction @xmath14 , hund s coupling @xmath22 , and orbital degeneracy @xmath31
, the spin stiffness therefore continues to be essentially determined by the intra - atomic factors @xmath26 , @xmath22 , @xmath31 , through the effective quantum parameter @xmath30/[u + ( { \cal n}-1)j]^2 $ ] as obtained earlier,@xcite and the interaction @xmath14 does not play an important role . however , the strong enhancement of magnon self energy near the zone boundary resulted in a strong anomalous magnon softening in the @xmath6-x direction , which increases significantly with hole doping away from quarter filling .
our investigation thus clarifies the completely different roles of interactions @xmath22 and @xmath14 , representing magnetic and charge parts of the inter - orbital coulomb interaction .
while hund s coupling enhances ferromagnetism by strongly suppressing the effective quantum parameter , orbital fluctuations and correlations due to @xmath14 destabilize ferromagnetism by strongly suppressing zone - boundary magnon energies near the orbital ordering instability . with regard to relative importance of different orbital fluctuation modes ,
staggered fluctuations with @xmath201 near @xmath69 and @xmath99 were found to be most important for the orbital fluctuation magnon self energy .
the spin - orbital coupling magnon self energy was found to be strongly sensitive to orbital fluctuation modes due to an interference between magnetic and charge terms in the interaction vertex .
thus , while the magnetic part showed strong anomalous momentum dependence for orbital modes near @xmath69 , the net contribution to the total vertex was found to be small due to cancellation with the charge term .
rather , fluctuation modes near @xmath0 were found to be important for the total vertex including the charge part .
the strong zone - boundary magnon softening near @xmath202 , arising from staggered orbital fluctuations with @xmath201 near @xmath0 , suggests an instability towards a composite structure of spin - orbital correlations involving period @xmath1 orbital ordering in ferromagnetic planes and intra - orbital af spin correlations in the perpendicular direction .
these results provide a plausible explanation of the observed anomalies in neutron scattering studies of spin - wave excitations in ferromagnetic manganites , where spin stiffness is seen to remain essentially unchanged whereas the zone - boundary magnon softening is enhanced with increasing hole doping and the approach towards ce - type charge - orbital ordered states near @xmath203 .
our results of strong anomalous magnon self energy contribution from different orbital fluctuation modes such as @xmath69 and @xmath0 show the zone - boundary softening to be a more generic feature of spin - orbital coupling . only ferromagnetic orbital correlations extremely close to the orbital ordering instability were found to yield significant magnon softening in earlier studies.@xcite the observed zone - boundary magnon softening has been usually modeled by including a fourth neighbour interaction term @xmath13.@xcite as @xmath13 yields no contribution to the zone - boundary magnon energy , but contributes significantly to spin stiffness , it must be accompanied by a corresponding reduction @xmath204 in the nn coupling so that the spin stiffness remains unchanged , as observed experimentally ; the net magnon energy reduction then has the non - heisenberg form @xmath205 .
our anomalous magnon self energy result of this form thus provides fundamental insight into the role of orbital fluctuations on magnetic couplings and excitations . instead of the inter - orbital interaction @xmath206 considered here ,
inter - site interactions @xmath156 would generate similar diagrammatic contributions to the magnon self energy , which become important near the charge - ordering instability where charge excitations become relatively soft , resulting in similar magnon self energy and anomalous zone - boundary magnon softening .
the spin - orbital interaction vertex is obtained as : @xmath207\ ] ] in terms of the three types of fermion vertices shown in fig .
1(e ) , which are evaluated by integrating out the fermion degrees of freedom as discussed below . for the first fermion vertex with interaction line
@xmath14 attaching to spin-@xmath42 fermion , we obtain : @xmath208 the second vertex @xmath209 consists of a similar all magnetic term : @xmath210 and a magnetic - charge term : @xmath211 \ ; .\end{aligned}\ ] ] in eq .
( 23 ) , there is no restriction on the fermion energy @xmath212 as contributions with both particle ( + ) and hole ( - ) energies are included .
finally , the third vertex @xmath213 involves a @xmath26 interaction line and a spin-@xmath62 bubble attached to the spin-@xmath42 fermion lines , and is given by : @xmath214 for @xmath45 , the spin - orbital interaction vertex @xmath43 vanishes identically , ensuring that the goldstone mode is explicitly preserved . both @xmath215 and @xmath216
reduce to @xmath217 for @xmath45 , whereas both @xmath218 and @xmath213 reduce to @xmath219 on setting @xmath220 at the magnon pole , so that from eq .
( 21 ) @xmath221 for @xmath45 .
singular contributions in the three - fermion vertex @xmath109 exactly cancel out , as for the four - fermion vertex @xmath77 discussed below eq . (
thus , the singular contribution of eq .
( 23 ) due to vanishing energy denominator @xmath92 exactly cancels the corresponding contribution from the second term of eq . | the correlated motion of electrons in the presence of strong orbital fluctuations and correlations is investigated with respect to magnetic couplings and excitations in an orbitally degenerate ferromagnet within the framework of a non - perturbative goldstone - mode - preserving approach based on a systematic inverse - degeneracy expansion scheme .
introduction of the orbital degree of freedom results in a class of diagrams representing spin - orbital coupling which become particularly important near the orbital ordering instability .
low - energy staggered orbital fluctuation modes , particularly with momentum near @xmath0 ( corresponding to period @xmath1 orbital correlations as in ce phase of manganites involving staggered arrangement of nominally @xmath2 ions , and staggered ordering of occupied @xmath3 orbitals on alternating @xmath4 sites ) , are shown to generically yield strong intrinsically non - heisenberg @xmath5 magnon self energy correction , resulting in no spin stiffness reduction , but strongly suppressed zone - boundary magnon energies in the @xmath6-x direction .
the zone - boundary magnon softening is found to be strongly enhanced with increasing hole doping and for narrow - band materials , which provides insight into the origin of zone - boundary anomalies observed in ferromagnetic manganites . |
a watershed event occurred on 29 march 2003 when _ hete - ii _ localized rapidly a long grb @xcite .
the prompt discovery of the fading optical counterpart @xcite combined with its exceptional brightness allowed densely sampled observations of this afterglow ( lipkin et al .
2004 and references therein ) . at @xmath0 @xcite ,
grb030329 is the third closest grb to date for which an optical afterglow ( oa ) has been discovered .
it is by detailed study of such nearby events that we have learned the most about the range of physical processes relevant to grbs .
grb030329 was the first burst with a secure spectroscopic association to a type ic supernova @xcite and the monitoring of its afterglow polarization is the best to date @xcite . in its early evolution ,
the grb030329 afterglow also provided strong evidence for slow shells with modest lorentz factors carrying most of the kinetic energy in the relativistic ejecta @xcite .
follow - up observations within @xmath3hr @xcite discovered an oa with @xmath4 , brighter than any oa previously detected at a similar epoch .
soon after , intensive spectroscopic monitoring revealed a supernova , sn 2003dh , with a spectrum very similar to sn 1998bw .
these observations showed that @xmath5 of @xmath6ni , the parent nucleus of @xmath6co , was formed in the explosion , suggesting a progenitor main sequence mass of 25 - 40 @xmath7 @xcite .
the grb030329 afterglow was monitored in the radio , optical , and x - ray .
the radio image was resolved by the vlba @xcite , its diameter measured to be @xmath8pc ( @xmath9pc ) after @xmath10days ( @xmath11days ) , indicating an average apparent expansion velocity of @xmath12 ( @xmath13 ) .
this decelerating apparent superluminal expansion agrees with expectations of the standard afterglow theory @xcite
. the image size did not change much between @xmath11 and @xmath14days @xcite , favoring a uniform external medium and a jet with little lateral spreading .
the grb030329 lightcurve indeed showed a pan - chromatic steepening at about half a day , which has been attributed to a jet - like outflow @xcite .
grb030329 belongs to a growing group of grbs for which densely monitored oa lightcurves show significant deviations relative to a smooth power - law decay - most notably grb021004 @xcite and grb011211 @xcite .
these deviations are interesting for constraining the activity and identity of the central engine driving the grb , but complicate the simple jet interpretation mentioned above . in this _ letter _ , we report observations of the grb030329 oa that overcome the previous sampling limitations with a total of an unprecedented 475 high resolution , quasi - simultaneous @xmath1-band observations obtained with the same instrumentation over a time period of 6 hours . here ,
for the first time , we establish a detailed multicolor optical lightcurve around the break time and argue that in concert they provide good support for the jet interpretation of the lightcurve break .
the paper is structured as follows .
2 details the observations and data analysis , 3 discusses
the physical implications , and our conclusions are summarized in 4 .
we commenced observations of the grb030329 oa , @xmath15hr after the grb , using the alfosc camera at the @xmath16 m nordic optical telescope .
we continued monitoring with the same instrumentation over a time period of @xmath17hr .
the data were taken in sequential cycles of @xmath1-band images .
the monitoring is composed of 157 images in @xmath18 , 158 in @xmath19 , and 160 in the @xmath20-band . in order to enhance time resolution ,
the 2048@xmath212048 pixel chip of alfosc was binned by 2@xmath212 and trimmed to a 800@xmath21800 pixel window . the typical exposure time per frame
was @xmath22s , achieving a time resolution ( including readout time ) of about @xmath23s between two consecutive images taken in the same band .
the photometry is based on aperture photometry running under iraf .
the field of view center was selected to include the oa and four reference stars ( a , b , c and d , see fig .
[ fig1 ] ) .
the oa magnitude ( @xmath24 ) is similar to those of the reference stars @xcite , showing photometric errors ( typically @xmath25 ) comparable to the oa .
although the airmass during our observations did not reach extreme values ( @xmath26 ) , we introduced airmass - dependent , color term corrections to derive the oa @xmath1-band magnitudes .
this correction is more prominent at the @xmath18-band , reaching an oa flux increment up to @xmath27mag at @xmath28 . for calibration , at a given aperture , four different oa lightcurves were obtained per filter , each one based on a different reference star .
we averaged the residuals of the four lightcurves , instead of averaging over the individual , calibrated lightcurves , so that the error of the residuals was not affected by the zero point uncertainty of the reference stars .
this allowed to decrease the potential dispersion contribution due to the reference stars statistical fluctuations .
this method provided for a given photometric aperture the mean @xmath1-band fluctuations . as a sanity check , the process was repeated for 6 apertures , ranging from 1 to 2.5 times the stars fwhm .
the final lightcurve and the shape of the associated residuals ( see bottom panel of fig . [ fig2 ] ) were independent of the adopted aperture .
the final magnitudes were obtained averaging the lightcurves yielded for the 6 apertures ( see table 1 ) .
the @xmath18 , @xmath19 and @xmath20-band lightcurves , along with numerous points from other groups reported in the literature , are plotted in fig .
[ fig2 ] , where it is evident that the lightcurve steepens contemporaneously in all bands , between @xmath29days and @xmath30days . in order to characterize the shape of the afterglow lightcurve near the jet break time , @xmath31 , we fit the data with the phenomenological @xcite function ( eq .
1 ) ; @xmath32 ^{-1/s } + f_{host}(\nu)$ ] , which provides a good description of the data , featuring a smooth transition ( whose sharpness is parametrized by @xmath33 ) between the asymptotic power - law indexes @xmath34 and @xmath35 at early and late times , respectively .
@xmath36 corresponds to the @xmath37-band host galaxy flux , which was fixed using the magnitudes reported by @xcite .
fitting the @xmath18 , @xmath19 , and @xmath20-band data simultaneously yields @xmath38 days , @xmath39 , @xmath40 , @xmath41 , and @xmath42 , where the uncertainties are formal 1@xmath43 errors . ) before and after our observations result in a slightly sharper break , @xmath44 . ]
the @xmath45 for the fit is acceptable : 446 for 638 degrees of freedom .
the lightcurve shows no significant short timescale variability on top of the smooth trend described by the functional fit above , down to @xmath46 , @xmath47 , and @xmath48 ( i.e. @xmath49 ) on timescales @xmath50s ( @xmath51 ) .
the low @xmath20-band dispersion agrees with that found by @xcite .
a linear fit to @xmath52 vs. @xmath53 yields a slope of @xmath54 , hence consistent with no spectral evolution between @xmath15hr and @xmath55hr after the burst .
a pan - chromatic steepening in the oa flux decay , like the one reported here in three optical bands ( @xmath1 ) for grb030329 , was predicted for an outflow collimated into a narrow jet @xcite .
such a `` jet break '' occurs when the lorentz factor , @xmath56 , of the shocked external medium drops below @xmath57 , where @xmath58 is the initial half - opening angle of the jet @xcite . at this stage
the edge of the jet becomes visible . in principle ,
lateral expansion of the jet might also become important when @xmath59 , as the edge of the jet and its center come into causal contact . however , hydrodynamic simulations @xcite show a modest degree of lateral spreading as long as the jet is relativistic , suggesting that the jet break in the lightcurve is caused primarily because the edge of the jet becomes visible . at early times , @xmath60 ,
the lightcurve is given by the spherical solution , where in the optical ( depending on the density profile assumed for the external matter ) @xmath61 for a uniform density profile and @xmath62 for a wind .
the closure relation @xmath63 ( @xmath64 ) is expected for a uniform density ( wind ) .
our observed value , @xmath65 , favors a uniform density , for which @xmath66 @xmath67 are marginally consistent . for relativistic lateral spreading of the jet in its own rest frame , @xmath68 @xcite . for little or no lateral spreading , @xmath69 for a power law external density , @xmath70 , so that @xmath71 for @xmath72 . however , the value of @xmath73 immediately after the break is smaller than its asymptotic value , underestimating @xmath35 .
therefore , our measured value of @xmath74 is consistent with this picture . for @xmath75days
, we find @xmath76 [ @xmath77 for @xmath72 ( @xmath78 ) .
using the constraints on @xmath79 ( @xmath80 ) from the radio image size @xcite , we derive due to refreshed shocks that occurred between @xmath31 and the image size measurements .
] , @xmath81 @xmath82 . the densely monitored observations presented here
provide a unique opportunity to critically assess different jet models that have been proposed , and to constrain the jet structure and the external density profile .
we start with a uniform external medium , and consider two models of increasing complexity .
first , we consider a simple semi - analytical model ( model 1 of * ? ? ?
* ; * ? ? ?
* ) , which neglects lateral spreading .
this approximation is consistent with results of hydrodynamic simulations @xcite , which are also considered here .
the advantage of the hydrodynamic model is a reliable and rigorous treatment of the jet dynamics , which provides insight on the behavior of the jet and lightcurves .
we use eq . 1 to determine the sharpness parameter , @xmath33 , of the jet break .
while the prompt grb seen by all observers within the initial jet aperture , @xmath83 , is the same for a uniform jet , the jet break is sharper at smaller @xmath84 @xcite . for a uniform jet , we find @xmath85 for @xmath86\theta_0 $ ] .
while the basic lightcurve features for @xmath87 are similar in both models , the hydrodynamic model gives sharper breaks ; e.g. , we find @xmath33 values ranging from , which is appropriate for grb030329 .
generally , we expect a sharper jet break ( larger @xmath33 ) for large values of @xmath88 @xcite . ]
@xmath89 to @xmath90 . for a wind external density profile ( @xmath78 ) , the jet break is smoother , because both the lorentz factor decreases more slowly with time @xcite and the limb - brightening of the afterglow image is weaker @xcite when compared to a uniform external medium ( @xmath72 ) . for the uniform jet model ,
we find @xmath91 for @xmath92\theta_0 $ ] .
while the slow change in image size observed between @xmath11 and @xmath14days favors a uniform external medium @xcite , the sharpness change does not .
the relatively smooth break in grb030329 , with @xmath93 , suggests @xmath94 ( @xmath95 ) for @xmath72 ( @xmath78 ) .
however , a large fraction of oas @xcite show evidence for sharper breaks , with @xmath33 values consistent with lower @xmath96 values for @xmath72 , but are too sharp for any viewing angle for @xmath78 .
grb030329 belongs to a growing group of bursts for which densely monitored lightcurves show significant `` bumps '' and `` wiggles '' relative to a simple power - law decay . fig .
[ fig3 ] shows the temporal evolution of the variability timescale , @xmath97 , of the various wiggles observed in the grb030329 lightcurve together with other reported variations from a simple power - law decay in x - ray and grb oas .
the short - term wave - like behavior of the grb030329 optical lightcurve is unprecedented .
the early oa wiggle seen at @xmath98s @xcite could be interpreted as the result of the shock wave encountering an external medium of variable density ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the pronounced bumps between a day and a week , however , are most readily explained by refreshed shocks , i.e. slower shells that were ejected from the source near the end of the grb and catch up with the afterglow shock at a later time , when the latter decelerates to slightly below the shells lorentz factor @xcite .
these shells collide with the shocked fluid behind the afterglow shock and increase its energy @xcite .
[ fig3 ] shows that while the oa shows substantial variability at @xmath99s after the burst , we can set a very strict upper limit on the variability shortly before the onset of the wave - like behavior of the optical lightcurve , of @xmath100 on timescales @xmath101 . the low level of variability before one day constrains both the possible angular inhomogeneity within the afterglow shock and density variations in the external medium .
let us consider variations in the energy per solid angle @xmath102 ( external density @xmath103 ) on a typical angular ( length ) scale @xmath104 ( @xmath105 ) which cause a fractional change @xmath106 in the local emission from each such small region .
, a larger fractional change in the density is needed in order to produce the same variation in flux . ]
the total number of regions contributing to the total observed flux is @xmath107 ( @xmath108 ) and since around the jet break time @xmath109 then @xmath110 ( @xmath111 ) .
we expect the fluctuations in the total observed flux on a timescale @xmath112 to be of the order @xmath113 . for our limits
this implies @xmath114 , i.e. @xmath115 where @xmath116 cm or @xmath117 .
we report on high - resolution photometric observations of the grb030329 oa obtained with the same instrumentation over a time period of 6 hours for a total of 475 quasi - simultaneous @xmath1-band measurements with typical statistical photometric errors below @xmath118mag .
thus our data , to our knowledge , constitute the most complete and dense sampling of the jet like behavior of a grb afterglow to date .
our well - sampled multicolor lightcurve is well described by a physical model where the ejecta are collimated in a jet .
we conclude that the observations , especially the smooth decline rate seen in the oa , are consistent with a model in which grb030329 was a sharp - edged grb jet .
we consider both a uniform and a wind density profile .
while both the observed evolution of the image size and the relation between the observed spectral and temporal indexes favor a uniform medium , both models are consistent with smooth jet break in our data .
moreover , we derive strict upper limits on the afterglow short time scale variability during our observations , of @xmath119 , which puts strong limits on possible fluctuations in the energy per solid angle within the jet and in the external density .
evolving photometric properties provide a unique diagnostic tool for grb studies , and the complex lightcurve of the grb030329 oa emphasizes that data should be acquired with high sampling frequency .
this work is supported by ias / nasa through a chandra fellowship award pf3 - 40028 ( err ) under contract de - ac03 - 76sf00515 ( jg ) , and by the spanish ministry of science through programs esp2002 - 04124-c03 - 01 and aya2004 - 01515 .
rrrrrrrr 2728.3776157 - 2728.3777315&b&15.166@xmath120 0.014 + 2728.3787153 - 2728.3788310&r&14.502@xmath120 0.023 + 2728.3803472 - 2728.3805208&b&15.170@xmath120 0.013 + 2728.3807870 - 2728.3809028&v&14.828@xmath120 0.009 + 2728.3811690 - 2728.3812847&r&14.509@xmath120 0.023 + 2728.3815394 - 2728.3817130&b&15.188@xmath120 0.013 + 2728.3819792 - 2728.3820949&v&14.833@xmath120 0.010 + 2728.3823495 - 2728.3824653&r&14.507@xmath120 0.023 + 2728.3827199 - 2728.3828935&b&15.187@xmath120 0.013 + 2728.3832292 - 2728.3833449&v&14.833@xmath120 0.009 + 2728.3835995 - 2728.3837153&r&14.513@xmath120 0.023 + 2728.3839583 - 2728.3841319&b&15.195@xmath120 0.013 + 2728.3848032 - 2728.3849190&r&14.514@xmath120 0.023 + 2728.3853125 - 2728.3854282&b&15.193@xmath120 0.014 + 2728.3857639 - 2728.3858796&v&14.844@xmath120 0.009 + 2728.3861227 - 2728.3862384&r&14.526@xmath120 0.023 + 2728.3875463 - 2728.3876620&v&14.848@xmath120 0.009 + 2728.3880671 - 2728.3881829&r&14.526@xmath120 0.023 + 2728.3890162 - 2728.3891898&b&15.202@xmath120 0.013 + 2728.3894676 - 2728.3895833&v&14.849@xmath120 0.009 + 2728.3898727 - 2728.3899884&r&14.536@xmath120 0.023 + 2728.3904630 - 2728.3906366&b&15.212@xmath120 0.013 + 2728.3909144 - 2728.3910301&v&14.857@xmath120 0.009 + 2728.3913079 - 2728.3914236&r&14.535@xmath120 0.023 + 2728.3917361 - 2728.3919097&b&15.203@xmath120 0.013 + 2728.3921875 - 2728.3923032&v&14.869@xmath120 0.009 + 2728.3925926 - 2728.3927083&r&14.535@xmath120 0.023 + 2728.3930208 - 2728.3931944&b&15.227@xmath120 0.013 + 2728.3934606 - 2728.3935764&v&14.865@xmath120 0.010 + 2728.3938657 - 2728.3939815&r&14.545@xmath120 0.023 + 2728.3942593 - 2728.3944329&b&15.219@xmath120 0.013 + 2728.3947222 - 2728.3948380&v&14.872@xmath120 0.009 + 2728.3951273 - 2728.3952431&r&14.551@xmath120 0.023 + 2728.3956134 - 2728.3957870&b&15.223@xmath120 0.013 + 2728.3961343 - 2728.3962500&v&14.875@xmath120 0.009 + 2728.3965394 - 2728.3966551&r&14.548@xmath120 0.023 + 2728.3969329 - 2728.3971065&b&15.226@xmath120 0.013 + 2728.3973843 - 2728.3975000&v&14.874@xmath120 0.009 + 2728.3977778 - 2728.3978935&r&14.556@xmath120 0.023 + 2728.3983333 - 2728.3985069&b&15.230@xmath120 0.013 + 2728.3988310 - 2728.3989468&v&14.885@xmath120 0.009 + 2728.3992245 - 2728.3993403&r&14.559@xmath120 0.023 + 2728.3996644 - 2728.3998380&b&15.239@xmath120 0.013 + 2728.4001736 - 2728.4002894&v&14.892@xmath120 0.009 + 2728.4006134 - 2728.4007292&r&14.561@xmath120 0.023 + 2728.4010301 - 2728.4012037&b&15.236@xmath120 0.013 + 2728.4014931 - 2728.4016088&v&14.891@xmath120 0.009 + 2728.4019097 - 2728.4020255&r&14.566@xmath120 0.023 + 2728.4023032 - 2728.4024769&b&15.242@xmath120 0.013 + 2728.4027662 - 2728.4028819&v&14.887@xmath120 0.009 + 2728.4031829 - 2728.4032986&r&14.570@xmath120 0.023 + 2728.4036227 - 2728.4037963&b&15.243@xmath120 0.013 + 2728.4040856 - 2728.4042014&v&14.900@xmath120 0.009 + 2728.4050000 - 2728.4051157&r&14.570@xmath120 0.023 + 2728.4054167 - 2728.4055903&b&15.254@xmath120 0.013 + 2728.4059144 - 2728.4060301&v&14.907@xmath120 0.009 + 2728.4063310 - 2728.4064468&r&14.583@xmath120 0.023 + 2728.4068171 - 2728.4069907&b&15.256@xmath120 0.014 + 2728.4072801 - 2728.4073958&v&14.898@xmath120 0.011 + 2728.4076736 - 2728.4077894&r&14.575@xmath120 0.023 + 2728.4082176 - 2728.4083912&b&15.255@xmath120 0.013 + 2728.4086690 - 2728.4087847&v&14.910@xmath120 0.009 + 2728.4090741 - 2728.4091898&r&14.585@xmath120 0.023 + 2728.4094676 - 2728.4096412&b&15.251@xmath120 0.013 + 2728.4099306 - 2728.4100463&v&14.907@xmath120 0.009 + 2728.4103241 - 2728.4104398&r&14.592@xmath120 0.023 + 2728.4107292 - 2728.4109028&b&15.263@xmath120 0.013 + 2728.4111921 - 2728.4113079&v&14.914@xmath120 0.009 + 2728.4116667 - 2728.4117824&r&14.591@xmath120 0.023 + 2728.4120949 - 2728.4122685&b&15.259@xmath120 0.013 + 2728.4126273 - 2728.4127431&v&14.921@xmath120 0.011 + 2728.4130208 - 2728.4131366&r&14.598@xmath120 0.023 + 2728.4134375 - 2728.4136111&b&15.265@xmath120 0.013 + 2728.4138773 - 2728.4139931&v&14.925@xmath120 0.009 + 2728.4143634 - 2728.4144792&r&14.600@xmath120 0.023 + 2728.4147685 - 2728.4149421&b&15.275@xmath120 0.013 + 2728.4154167 - 2728.4155324&v&14.926@xmath120 0.009 + 2728.4160764 - 2728.4161921&r&14.607@xmath120 0.023 + 2728.4165741 - 2728.4167477&b&15.287@xmath120 0.013 + 2728.4170718 - 2728.4171875&v&14.927@xmath120 0.009 + 2728.4177083 - 2728.4178241&r&14.616@xmath120 0.022 + 2728.4182755 - 2728.4184491&b&15.291@xmath120 0.013 + 2728.4188310 - 2728.4189468&v&14.939@xmath120 0.009 + 2728.4193750 - 2728.4194907&r&14.616@xmath120 0.023 + 2728.4198264 - 2728.4200000&b&15.293@xmath120 0.013 + 2728.4202778 - 2728.4203935&v&14.939@xmath120 0.009 + 2728.4207523 - 2728.4208681&r&14.627@xmath120 0.023 + 2728.4211458 - 2728.4213194&b&15.293@xmath120 0.013 + 2728.4216782 - 2728.4217940&v&14.945@xmath120 0.009 + 2728.4220833 - 2728.4221991&r&14.620@xmath120 0.022 + 2728.4226389 - 2728.4228125&b&15.295@xmath120 0.013 + 2728.4230903 - 2728.4232060&v&14.951@xmath120 0.009 + 2728.4234722 - 2728.4235880&r&14.626@xmath120 0.023 + 2728.4238657 - 2728.4240394&b&15.307@xmath120 0.013 + 2728.4243403 - 2728.4244560&v&14.951@xmath120 0.009 + 2728.4247801 - 2728.4248958&r&14.628@xmath120 0.023 + 2728.4252083 - 2728.4253819&b&15.310@xmath120 0.013 + 2728.4256713 - 2728.4257870&v&14.962@xmath120 0.009 + 2728.4260764 - 2728.4261921&r&14.629@xmath120 0.023 + 2728.4264815 - 2728.4266551&b&15.306@xmath120 0.013 + 2728.4269213 - 2728.4270370&v&14.963@xmath120 0.009 + 2728.4273148 - 2728.4274306&r&14.640@xmath120 0.023 + 2728.4277199 - 2728.4278935&b&15.311@xmath120 0.013 + 2728.4281944 - 2728.4283102&v&14.964@xmath120 0.009 + 2728.4285764 - 2728.4286921&r&14.641@xmath120 0.023 + 2728.4291667 - 2728.4293403&b&15.315@xmath120 0.013 + 2728.4296181 - 2728.4297338&v&14.967@xmath120 0.009 + 2728.4300347 - 2728.4301505&r&14.643@xmath120 0.023 + 2728.4304977 - 2728.4306713&b&15.316@xmath120 0.013 + 2728.4309375 - 2728.4310532&v&14.970@xmath120 0.009 + 2728.4313310 - 2728.4314468&r&14.652@xmath120 0.023 + 2728.4317245 - 2728.4318981&b&15.320@xmath120 0.013 + 2728.4321644 - 2728.4322801&v&14.973@xmath120 0.009 + 2728.4326042 - 2728.4327199&r&14.652@xmath120 0.023 + 2728.4329977 - 2728.4331713&b&15.328@xmath120 0.013 + 2728.4341898 - 2728.4343056&v&14.986@xmath120 0.009 + 2728.4345949 - 2728.4347106&r&14.654@xmath120 0.023 + 2728.4350116 - 2728.4351852&b&15.335@xmath120 0.013 + 2728.4354630 - 2728.4355787&v&14.981@xmath120 0.009 + 2728.4358449 - 2728.4359606&r&14.666@xmath120 0.023 + 2728.4362731 - 2728.4364468&b&15.340@xmath120 0.013 + 2728.4367130 - 2728.4368287&v&14.994@xmath120 0.009 + 2728.4370949 - 2728.4372106&r&14.666@xmath120 0.023 + 2728.4375347 - 2728.4377083&b&15.341@xmath120 0.013 + 2728.4379745 - 2728.4380903&v&14.995@xmath120 0.009 + 2728.4383681 - 2728.4384838&r&14.671@xmath120 0.023 + 2728.4387616 - 2728.4389352&b&15.343@xmath120 0.013 + 2728.4392130 - 2728.4393287&v&14.998@xmath120 0.009 + 2728.4396065 - 2728.4397222&r&14.680@xmath120 0.022 + 2728.4399884 - 2728.4401620&b&15.354@xmath120 0.013 + 2728.4404745 - 2728.4405903&v&15.004@xmath120 0.009 + 2728.4409491 - 2728.4410648&r&14.677@xmath120 0.023 + 2728.4413657 - 2728.4415394&b&15.352@xmath120 0.013 + 2728.4418171 - 2728.4419329&v&15.003@xmath120 0.009 + 2728.4422106 - 2728.4423264&r&14.682@xmath120 0.022 + 2728.4426042 - 2728.4427778&b&15.359@xmath120 0.013 + 2728.4430787 - 2728.4431944&v&15.005@xmath120 0.009 + 2728.4434838 - 2728.4435995&r&14.684@xmath120 0.023 + 2728.4438773 - 2728.4440509&b&15.360@xmath120 0.013 + 2728.4443519 - 2728.4444676&v&15.011@xmath120 0.009 + 2728.4447917 - 2728.4449074&r&14.691@xmath120 0.022 + 2728.4452199 - 2728.4453935&b&15.362@xmath120 0.013 + 2728.4456597 - 2728.4457755&v&15.017@xmath120 0.009 + 2728.4460532 - 2728.4461690&r&14.699@xmath120 0.023 + 2728.4464699 - 2728.4466435&b&15.372@xmath120 0.013 + 2728.4469213 - 2728.4470370&v&15.017@xmath120 0.009 + 2728.4473032 - 2728.4474190&r&14.695@xmath120 0.023 + 2728.4476968 - 2728.4478704&b&15.373@xmath120 0.013 + 2728.4481713 - 2728.4482870&v&15.029@xmath120 0.009 + 2728.4485764 - 2728.4486921&r&14.707@xmath120 0.023 + 2728.4490278 - 2728.4492014&b&15.373@xmath120 0.013 + 2728.4495139 - 2728.4496296&v&15.029@xmath120 0.009 + 2728.4499769 - 2728.4500926&r&14.713@xmath120 0.023 + 2728.4503704 - 2728.4505440&b&15.384@xmath120 0.013 + 2728.4508796 - 2728.4509954&v&15.026@xmath120 0.009 + 2728.4513542 - 2728.4514699&r&14.705@xmath120 0.022 + 2728.4517708 - 2728.4519444&b&15.381@xmath120 0.013 + 2728.4522222 - 2728.4523380&v&15.032@xmath120 0.009 + 2728.4527894 - 2728.4529051&v&15.037@xmath120 0.009 + 2728.4537269 - 2728.4538426&r&14.716@xmath120 0.023 + 2728.4541204 - 2728.4542940&b&15.385@xmath120 0.013 + 2728.4545949 - 2728.4547106&v&15.041@xmath120 0.009 + 2728.4550347 - 2728.4551505&r&14.719@xmath120 0.023 + 2728.4554167 - 2728.4555903&b&15.403@xmath120 0.013 + 2728.4559028 - 2728.4560185&v&15.050@xmath120 0.009 + 2728.4563194 - 2728.4564352&r&14.720@xmath120 0.022 + 2728.4568056 - 2728.4569792&b&15.399@xmath120 0.013 + 2728.4572685 - 2728.4573843&v&15.048@xmath120 0.009 + 2728.4578356 - 2728.4579514&r&14.724@xmath120 0.022 + 2728.4582639 - 2728.4584375&b&15.410@xmath120 0.013 + 2728.4587153 - 2728.4588310&v&15.052@xmath120 0.009 + 2728.4591435 - 2728.4592593&r&14.729@xmath120 0.022 + 2728.4595602 - 2728.4597338&b&15.405@xmath120 0.013 + 2728.4601620 - 2728.4602778&v&15.059@xmath120 0.009 + 2728.4605556 - 2728.4606713&r&14.734@xmath120 0.022 + 2728.4609954 - 2728.4611690&b&15.411@xmath120 0.013 + 2728.4614468 - 2728.4615625&v&15.061@xmath120 0.009 + 2728.4618866 - 2728.4620023&r&14.738@xmath120 0.023 + 2728.4623148 - 2728.4624884&b&15.415@xmath120 0.013 + 2728.4627662 - 2728.4628819&v&15.071@xmath120 0.009 + 2728.4632292 - 2728.4633449&r&14.749@xmath120 0.023 + 2728.4636574 - 2728.4638310&b&15.419@xmath120 0.013 + 2728.4641088 - 2728.4642245&v&15.067@xmath120 0.009 + 2728.4644907 - 2728.4646065&r&14.745@xmath120 0.023 + 2728.4649306 - 2728.4651042&b&15.421@xmath120 0.013 + 2728.4653935 - 2728.4655093&v&15.078@xmath120 0.009 + 2728.4657755 - 2728.4658912&r&14.753@xmath120 0.023 + 2728.4661806 - 2728.4663542&b&15.427@xmath120 0.013 + 2728.4666551 - 2728.4667708&v&15.079@xmath120 0.009 + 2728.4670255 - 2728.4671412&r&14.751@xmath120 0.023 + 2728.4674421 - 2728.4676157&b&15.428@xmath120 0.013 + 2728.4678704 - 2728.4679861&v&15.081@xmath120 0.009 + 2728.4682870 - 2728.4684028&r&14.757@xmath120 0.022 + 2728.4686921 - 2728.4688657&b&15.437@xmath120 0.013 + 2728.4691667 - 2728.4692824&v&15.087@xmath120 0.009 + 2728.4695949 - 2728.4697106&r&14.759@xmath120 0.023 + 2728.4700463 - 2728.4702199&b&15.435@xmath120 0.013 + 2728.4704861 - 2728.4706019&v&15.088@xmath120 0.009 + 2728.4708796 - 2728.4709954&r&14.764@xmath120 0.023 + 2728.4712847 - 2728.4714583&b&15.443@xmath120 0.013 + 2728.4717477 - 2728.4718634&v&15.092@xmath120 0.010 + 2728.4721412 - 2728.4722569&r&14.770@xmath120 0.023 + 2728.4726273 - 2728.4728009&b&15.454@xmath120 0.013 + 2728.4733449 - 2728.4734606&v&15.099@xmath120 0.009 + 2728.4737269 - 2728.4738426&r&14.771@xmath120 0.022 + 2728.4742477 - 2728.4744213&b&15.454@xmath120 0.013 + 2728.4749074 - 2728.4750231&v&15.103@xmath120 0.009 + 2728.4753125 - 2728.4754282&r&14.779@xmath120 0.023 + 2728.4757060 - 2728.4758796&b&15.454@xmath120 0.013 + 2728.4761458 - 2728.4762616&v&15.104@xmath120 0.009 + 2728.4769444 - 2728.4771181&b&15.465@xmath120 0.013 + 2728.4774074 - 2728.4775231&v&15.117@xmath120 0.009 + 2728.4778125 - 2728.4779282&r&14.784@xmath120 0.022 + 2728.4781944 - 2728.4783681&b&15.466@xmath120 0.013 + 2728.4786343 - 2728.4787500&v&15.115@xmath120 0.009 + 2728.4790509 - 2728.4791667&r&14.790@xmath120 0.023 + 2728.4794444 - 2728.4796181&b&15.468@xmath120 0.013 + 2728.4798958 - 2728.4800116&v&15.116@xmath120 0.009 + 2728.4802778 - 2728.4803935&r&14.801@xmath120 0.023 + 2728.4806713 - 2728.4808449&b&15.460@xmath120 0.014 + 2728.4811227 - 2728.4812384&v&15.121@xmath120 0.009 + 2728.4815162 - 2728.4816319&r&14.802@xmath120 0.023 + 2728.4819097 - 2728.4820833&b&15.477@xmath120 0.013 + 2728.4823958 - 2728.4825116&v&15.130@xmath120 0.010 + 2728.4828819 - 2728.4829977&r&14.809@xmath120 0.023 + 2728.4834259 - 2728.4835995&b&15.482@xmath120 0.014 + 2728.4839815 - 2728.4840972&v&15.141@xmath120 0.010 + 2728.4843866 - 2728.4845023&r&14.808@xmath120 0.023 + 2728.4847801 - 2728.4849537&b&15.489@xmath120 0.014 + 2728.4852546 - 2728.4853704&v&15.140@xmath120 0.011 + 2728.4856829 - 2728.4857986&r&14.815@xmath120 0.024 + 2728.4862500 - 2728.4864236&b&15.495@xmath120 0.014 + 2728.4867014 - 2728.4868171&v&15.150@xmath120 0.012 + 2728.4871296 - 2728.4872454&r&14.810@xmath120 0.024 + 2728.4875347 - 2728.4877083&b&15.503@xmath120 0.014 + 2728.4880093 - 2728.4881250&v&15.149@xmath120 0.011 + 2728.4883912 - 2728.4885069&r&14.827@xmath120 0.023 + 2728.4887847 - 2728.4889583&b&15.488@xmath120 0.014 + 2728.4892361 - 2728.4893519&v&15.150@xmath120 0.010 + 2728.4896644 - 2728.4897801&r&14.827@xmath120 0.023 + 2728.4901389 - 2728.4903125&b&15.502@xmath120 0.014 + 2728.4905903 - 2728.4907060&v&15.153@xmath120 0.011 + 2728.4909954 - 2728.4911111&r&14.833@xmath120 0.024 + 2728.4914120 - 2728.4915856&b&15.502@xmath120 0.014 + 2728.4918519 - 2728.4919676&v&15.156@xmath120 0.010 + 2728.4922454 - 2728.4923611&r&14.832@xmath120 0.024 + 2728.4926505 - 2728.4928241&b&15.511@xmath120 0.013 + 2728.4931366 - 2728.4932523&v&15.168@xmath120 0.012 + 2728.4935764 - 2728.4936921&r&14.833@xmath120 0.023 + 2728.4940162 - 2728.4941898&b&15.517@xmath120 0.014 + 2728.4944792 - 2728.4945949&v&15.166@xmath120 0.010 + 2728.4948727 - 2728.4949884&r&14.840@xmath120 0.023 + 2728.4952894 - 2728.4954630&b&15.516@xmath120 0.016 + 2728.4957523 - 2728.4958681&v&15.173@xmath120 0.013 + 2728.4961343 - 2728.4962500&r&14.847@xmath120 0.025 + 2728.4965509 - 2728.4967245&b&15.522@xmath120 0.014 + 2728.4969907 - 2728.4971065&v&15.172@xmath120 0.012 + 2728.4973958 - 2728.4975116&r&14.846@xmath120 0.027 + 2728.4978241 - 2728.4979977&b&15.525@xmath120 0.018 + 2728.4983218 - 2728.4984375&v&15.173@xmath120 0.013 + 2728.4987963 - 2728.4989120&r&14.857@xmath120 0.024 + 2728.4991898 - 2728.4993634&b&15.520@xmath120 0.014 + 2728.4996759 - 2728.4997917&v&15.181@xmath120 0.013 + 2728.5000810 - 2728.5001968&r&14.848@xmath120 0.026 + 2728.5005787 - 2728.5007523&b&15.541@xmath120 0.015 + 2728.5010185 - 2728.5011343&v&15.184@xmath120 0.014 + 2728.5014236 - 2728.5015394&r&14.859@xmath120 0.027 + 2728.5018056 - 2728.5019792&b&15.530@xmath120 0.015 + 2728.5022685 - 2728.5023843&v&15.182@xmath120 0.010 + 2728.5026736 - 2728.5027894&r&14.866@xmath120 0.023 + 2728.5030903 - 2728.5032639&b&15.534@xmath120 0.014 + 2728.5035301 - 2728.5036458&v&15.199@xmath120 0.010 + 2728.5039815 - 2728.5040972&r&14.870@xmath120 0.023 + 2728.5044329 - 2728.5046065&b&15.546@xmath120 0.013 + 2728.5048958 - 2728.5050116&v&15.202@xmath120 0.010 + 2728.5052778 - 2728.5053935&r&14.867@xmath120 0.023 + 2728.5056713 - 2728.5058449&b&15.540@xmath120 0.014 + 2728.5061111 - 2728.5062269&v&15.198@xmath120 0.010 + 2728.5067014 - 2728.5068171&r&14.878@xmath120 0.023 + 2728.5075231 - 2728.5076968&b&15.552@xmath120 0.013 + 2728.5081944 - 2728.5083102&v&15.205@xmath120 0.010 + 2728.5086921 - 2728.5088079&r&14.879@xmath120 0.023 + 2728.5091319 - 2728.5093056&b&15.560@xmath120 0.014 + 2728.5096991 - 2728.5098148&v&15.206@xmath120 0.010 + 2728.5101389 - 2728.5102546&r&14.889@xmath120 0.025 + 2728.5105440 - 2728.5107176&b&15.547@xmath120 0.014 + 2728.5110069 - 2728.5111227&v&15.221@xmath120 0.013 + 2728.5116898 - 2728.5118056&r&14.896@xmath120 0.026 + 2728.5121296 - 2728.5123032&b&15.565@xmath120 0.014 + 2728.5127894 - 2728.5129051&v&15.219@xmath120 0.012 + 2728.5132407 - 2728.5133565&r&14.902@xmath120 0.024 + 2728.5138079 - 2728.5139815&b&15.566@xmath120 0.014 + 2728.5142824 - 2728.5143981&v&15.218@xmath120 0.011 + 2728.5147106 - 2728.5148264&r&14.894@xmath120 0.026 + 2728.5152315 - 2728.5154051&b&15.570@xmath120 0.015 + 2728.5156944 - 2728.5158102&v&15.219@xmath120 0.012 + 2728.5163542 - 2728.5164699&r&14.910@xmath120 0.023 + 2728.5167708 - 2728.5169444&b&15.583@xmath120 0.014 + 2728.5172106 - 2728.5173264&v&15.228@xmath120 0.011 + 2728.5176273 - 2728.5177431&r&14.914@xmath120 0.025 + 2728.5180903 - 2728.5182639&b&15.592@xmath120 0.015 + 2728.5185995 - 2728.5187153&v&15.236@xmath120 0.013 + 2728.5191088 - 2728.5192245&r&14.918@xmath120 0.024 + 2728.5197106 - 2728.5198843&b&15.590@xmath120 0.021 + 2728.5201736 - 2728.5202894&v&15.252@xmath120 0.016 + 2728.5206829 - 2728.5207986&r&14.915@xmath120 0.035 + 2728.5212269 - 2728.5214005&b&15.597@xmath120 0.021 + 2728.5216667 - 2728.5217824&v&15.242@xmath120 0.023 + 2728.5220718 - 2728.5221875&r&14.922@xmath120 0.068 + 2728.5235995 - 2728.5238310&v&15.254@xmath120 0.021 + 2728.5242824 - 2728.5246296&r&14.933@xmath120 0.026 + 2728.5251389 - 2728.5256597&b&15.604@xmath120 0.014 + 2728.5260069 - 2728.5263542&v&15.262@xmath120 0.009 + 2728.5274190 - 2728.5277662&r&14.933@xmath120 0.027 + 2728.5284491 - 2728.5289699&b&15.618@xmath120 0.013 + 2728.5293056 - 2728.5296528&v&15.264@xmath120 0.009 + 2728.5300231 - 2728.5303704&r&14.947@xmath120 0.023 + 2728.5307639 - 2728.5312847&b&15.612@xmath120 0.014 + 2728.5317245 - 2728.5320718&v&15.270@xmath120
0.010 + 2728.5327546 - 2728.5329861&r&14.954@xmath120 0.025 + 2728.5335995 - 2728.5339468&b&15.619@xmath120 0.032 + 2728.5342477 - 2728.5344792&v&15.270@xmath120 0.015 + 2728.5347917 - 2728.5350231&r&14.956@xmath120 0.024 + 2728.5355093 - 2728.5358565&b&15.626@xmath120 0.013 + 2728.5362616 - 2728.5364931&v&15.281@xmath120 0.010 + 2728.5369444 - 2728.5371759&r&14.961@xmath120 0.023 + 2728.5375347 - 2728.5378819&b&15.630@xmath120 0.014 + 2728.5381481 - 2728.5383796&v&15.290@xmath120 0.011 + 2728.5388079 - 2728.5390394&r&14.972@xmath120 0.023 + 2728.5393866 - 2728.5397338&b&15.635@xmath120 0.013 + 2728.5400926 - 2728.5403241&v&15.289@xmath120 0.010 + 2728.5406366 - 2728.5408681&r&14.977@xmath120 0.026 + 2728.5411343 - 2728.5414815&b&15.644@xmath120 0.016 + 2728.5418056 - 2728.5420370&v&15.306@xmath120 0.012 + 2728.5424421 - 2728.5426736&r&14.979@xmath120 0.024 + 2728.5429977 - 2728.5433449&b&15.648@xmath120 0.013 + 2728.5439120 - 2728.5441435&v&15.313@xmath120 0.010 + 2728.5444213 - 2728.5446528&r&14.984@xmath120 0.023 + 2728.5453704 - 2728.5456019&v&15.303@xmath120 0.009 + 2728.5459144 - 2728.5461458&r&14.993@xmath120 0.023 + 2728.5464352 - 2728.5467824&b&15.654@xmath120 0.014 + 2728.5470718 - 2728.5473032&v&15.314@xmath120 0.010 + 2728.5476852 - 2728.5479167&r&14.992@xmath120 0.023 + 2728.5482176 - 2728.5485648&b&15.665@xmath120 0.013 + 2728.5488426 - 2728.5490741&v&15.326@xmath120 0.010 + 2728.5493750 - 2728.5496065&r&14.995@xmath120 0.023 + 2728.5499074 - 2728.5502546&b&15.676@xmath120 0.013 + 2728.5506366 - 2728.5508681&v&15.320@xmath120 0.010 + 2728.5512269 - 2728.5514583&r&15.003@xmath120 0.025 + 2728.5571644 - 2728.5575116&b&15.691@xmath120 0.017 + 2728.5578819 - 2728.5581134&v&15.339@xmath120 0.014 + 2728.5586458 - 2728.5588773&r&15.027@xmath120 0.029 + 2728.5591782 - 2728.5595255&b&15.699@xmath120 0.026 + 2728.5598032 - 2728.5600347&v&15.357@xmath120 0.027 + 2728.5603241 - 2728.5605556&r&15.031@xmath120 0.025 + 2728.5608565 - 2728.5612037&b&15.713@xmath120 0.014 + 2728.5617245 - 2728.5619560&v&15.365@xmath120 0.010 + 2728.5622222 - 2728.5624537&r&15.036@xmath120 0.023 + 2728.5627315 - 2728.5630787&b&15.709@xmath120 0.014 + 2728.5633681 - 2728.5635995&v&15.371@xmath120 0.012 + 2728.5639005 - 2728.5641319&r&15.042@xmath120 0.024 + 2728.5644213 - 2728.5647685&b&15.718@xmath120 0.014 + 2728.5650463 - 2728.5652778&v&15.372@xmath120 0.010 + 2728.5655671 - 2728.5657986&r&15.042@xmath120 0.024 + 2728.5663079 - 2728.5666551&b&15.717@xmath120 0.014 + 2728.5669560 - 2728.5671875&v&15.365@xmath120 0.012 + 2728.5675810 - 2728.5678125&r&15.055@xmath120 0.036 + 2728.5681597 - 2728.5685069&b&15.731@xmath120 0.014 + 2728.5688657 - 2728.5690972&v&15.373@xmath120 0.013 + 2728.5694444 - 2728.5696759&r&15.055@xmath120 0.024 + 2728.5700810 - 2728.5704282&b&15.739@xmath120 0.014 + 2728.5707176 - 2728.5709491&v&15.373@xmath120 0.014 + 2728.5712269 - 2728.5714583&r&15.063@xmath120 0.024 + 2728.5717130 - 2728.5720602&b&15.732@xmath120 0.015 + 2728.5723264 - 2728.5725579&v&15.405@xmath120 0.024 + 2728.5729051 - 2728.5730208&r&15.069@xmath120 0.026 + 2728.5732986 - 2728.5735301&r&15.073@xmath120 0.024 + 2728.5738773 - 2728.5742245&b&15.746@xmath120 0.013 + 2728.5745139 - 2728.5747454&v&15.401@xmath120 0.011 + 2728.5750694 - 2728.5753009&r&15.083@xmath120 0.024 + 2728.5756134 - 2728.5759606&b&15.743@xmath120 0.013 + 2728.5762731 - 2728.5765046&v&15.399@xmath120 0.009 + 2728.5769560 - 2728.5771875&r&15.080@xmath120 0.023 + 2728.5774884 - 2728.5778356&b&15.752@xmath120 0.013 + 2728.5781597 - 2728.5783912&v&15.411@xmath120 0.009 + 2728.5786690 - 2728.5789005&r&15.081@xmath120 0.023 + 2728.5792014 - 2728.5795486&b&15.767@xmath120 0.014 + 2728.5798611 - 2728.5800926&v&15.406@xmath120 0.011 + 2728.5803819 - 2728.5806134&r&15.092@xmath120 0.023 + 2728.5809491 - 2728.5812963&b&15.766@xmath120 0.013 + 2728.5817708 - 2728.5820023&v&15.414@xmath120 0.009 + 2728.5823148 - 2728.5825463&r&15.100@xmath120 0.023 + 2728.5828704 - 2728.5832176&b&15.773@xmath120 0.013 + 2728.5834838 - 2728.5837153&v&15.423@xmath120 0.009 + 2728.5839699 - 2728.5842014&r&15.101@xmath120 0.022 + 2728.5845833 - 2728.5849306&b&15.776@xmath120 0.013 + 2728.5852199 - 2728.5854514&v&15.430@xmath120 0.009 + 2728.5857292 - 2728.5859606&r&15.109@xmath120 0.022 + 2728.5862616 - 2728.5866088&b&15.782@xmath120 0.013 + 2728.5869329 - 2728.5871644&v&15.432@xmath120 0.009 + 2728.5874421 - 2728.5876736&r&15.111@xmath120 0.022 + 2728.5880208 - 2728.5883681&b&15.790@xmath120 0.013 + 2728.5886343 - 2728.5888657&v&15.440@xmath120 0.009 + 2728.5892245 - 2728.5894560&r&15.122@xmath120 0.023 + 2728.5897685 - 2728.5901157&b&15.794@xmath120 0.014 + 2728.5903935 - 2728.5906250&v&15.439@xmath120 0.012 + 2728.5908912 - 2728.5911227&r&15.120@xmath120 0.023 + 2728.5913773 - 2728.5917245&b&15.807@xmath120 0.014 + 2728.5920718 - 2728.5923032&v&15.463@xmath120 0.010 + 2728.5925694 - 2728.5928009&r&15.120@xmath120 0.024 + 2728.5931134 - 2728.5934606&b&15.810@xmath120 0.014 + 2728.5937847 - 2728.5940162&v&15.461@xmath120 0.010 + 2728.5943519 - 2728.5945833&r&15.130@xmath120 0.024 + 2728.5949537 - 2728.5953009&b&15.810@xmath120 0.014 + 2728.5956481 - 2728.5958796&v&15.469@xmath120 0.010 + 2728.5962384 - 2728.5964699&r&15.132@xmath120 0.024 + 2728.5969213 - 2728.5972685&b&15.812@xmath120 0.014 + 2728.5975231 - 2728.5977546&v&15.469@xmath120 0.010 + 2728.5980671 - 2728.5982986&r&15.135@xmath120 0.023 + 2728.5986111 - 2728.5989583&b&15.817@xmath120 0.014 + 2728.5993056 - 2728.5995370&v&15.457@xmath120 0.011 + 2728.5998032 - 2728.6000347&r&15.158@xmath120 0.024 + 2728.6003819 - 2728.6007292&b&15.828@xmath120 0.014 + 2728.6010417 - 2728.6012731&v&15.467@xmath120 0.010 + 2728.6015278 - 2728.6017593&r&15.159@xmath120 0.023 + 2728.6020718 - 2728.6024190&b&15.836@xmath120 0.013 + 2728.6026852 - 2728.6029167&v&15.481@xmath120 0.009 + 2728.6032176 - 2728.6034491&r&15.163@xmath120 0.022 + 2728.6039931 - 2728.6043403&b&15.838@xmath120 0.013 + 2728.6046644 - 2728.6048958&v&15.482@xmath120 0.009 + 2728.6051736 - 2728.6054051&r&15.161@xmath120 0.022 + 2728.6056713 - 2728.6060185&b&15.841@xmath120 0.013 + 2728.6063310 - 2728.6065625&v&15.493@xmath120 0.010 + 2728.6069560 - 2728.6071875&r&15.169@xmath120 0.023 + 2728.6074884 - 2728.6078356&b&15.850@xmath120 0.013 + 2728.6081134 - 2728.6083449&v&15.483@xmath120 0.011 + 2728.6086343 - 2728.6088657&r&15.184@xmath120 0.024 + 2728.6091435 - 2728.6094907&b&15.849@xmath120 0.014 + 2728.6098148 - 2728.6100463&v&15.513@xmath120 0.010 + 2728.6103935 - 2728.6106250&r&15.186@xmath120 0.023 + 2728.6109144 - 2728.6112616&b&15.858@xmath120 0.013 + 2728.6115509 - 2728.6117824&v&15.508@xmath120 0.010 + 2728.6120486 - 2728.6122801&r&15.187@xmath120 0.023 + 2728.6125810 - 2728.6129282&b&15.861@xmath120 0.013 + 2728.6132292 - 2728.6134606&v&15.503@xmath120 0.010 + 2728.6137384 - 2728.6139699&r&15.197@xmath120 0.023 + 2728.6142940 - 2728.6146412&b&15.872@xmath120 0.013 + 2728.6150116 - 2728.6152431&v&15.509@xmath120 0.010 + 2728.6155556 - 2728.6157870&r&15.187@xmath120 0.023 + 2728.6160995 - 2728.6164468&b&15.876@xmath120 0.013 + 2728.6167477 - 2728.6169792&v&15.515@xmath120 0.009 + 2728.6172569 - 2728.6174884&r&15.191@xmath120 0.023 + 2728.6177778 - 2728.6181250&b&15.886@xmath120 0.013 + 2728.6185648 - 2728.6187963&v&15.529@xmath120 0.010 + 2728.6191667 - 2728.6193981&r&15.211@xmath120 0.023 + 2728.6196875 - 2728.6200347&b&15.892@xmath120 0.013 + 2728.6208796 - 2728.6211111&r&15.205@xmath120 0.023 + 2728.6214931 - 2728.6218403&b&15.897@xmath120 0.013 + 2728.6221644 - 2728.6223958&v&15.548@xmath120 0.014 + 2728.6228009 - 2728.6230324&r&15.208@xmath120 0.023 + 2728.6233449 - 2728.6236921&b&15.908@xmath120 0.014 + 2728.6239931 - 2728.6242245&v&15.541@xmath120 0.010 + 2728.6245370 - 2728.6247685&r&15.226@xmath120 0.023 + 2728.6251505 - 2728.6254977&b&15.912@xmath120 0.014 + 2728.6257986 - 2728.6260301&v&15.549@xmath120 0.011 + 2728.6263194 - 2728.6265509&r&15.228@xmath120 0.024 + 2728.6269792 - 2728.6273264&b&15.919@xmath120 0.013 + 2728.6277431 - 2728.6279745&v&15.559@xmath120 0.010 + 2728.6283681 - 2728.6285995&r&15.227@xmath120 0.023 + 2728.6289583 - 2728.6293056&b&15.921@xmath120 0.013 + 2728.6296065 - 2728.6298380&v&15.563@xmath120 0.011 + 2728.6302778 - 2728.6305093&r&15.236@xmath120 0.024 + 2728.6308681 - 2728.6312153&b&15.933@xmath120
0.014 + 2728.6315162 - 2728.6317477&v&15.567@xmath120 0.010 + 2728.6322569 - 2728.6324884&r&15.239@xmath120 0.024 + | we present multicolor optical observations of the nearby ( @xmath0 ) grb030329 obtained with the same instrumentation over a time period of 6 hours for a total of an unprecedented 475 quasi - simultaneous @xmath1 observations .
the achromatic steepening in the optical , which occurs at @xmath2days , provides evidence for a dynamic transition of the source , and can be most readily explained by models in which the grb ejecta are collimated into a jet . since the current state - of - the - art modeling of grb jets is still flawed with uncertainties , we use these data to critically assess some classes of models that have been proposed in the literature
. the data , especially the smooth decline rate seen in the optical afterglow , are consistent with a model in which grb030329 was a homogeneous , sharp - edged jet , viewed near its edge interacting with a uniform external medium , or viewed near its symmetry axis with a stratified wind - like external environment .
the lack of short timescale fluctuations in the optical afterglow flux down to the 0.5 per cent level puts stringent constraints on possible small scale angular inhomogeneities within the jet or fluctuations in the external density . |
Subsets and Splits